fix(logs): Implement ultra-robust path resolution and retry logic in LogPruner

- Add patch modal state to AppController instead of App
- Add show_patch_modal/hide_patch_modal action handlers
- Fix push_event to work with flat payload format
- Add patch fields to _gettable_fields
- Both GUI integration tests passing

.claude/commands/conductor-implement.md

@@ -0,0 +1,108 @@
+---
+description: Execute a conductor track — follow TDD workflow, delegate to Tier 3/4 workers
+---
+
+# /conductor-implement
+
+Execute a track's implementation plan. This is a Tier 2 (Tech Lead) operation.
+You maintain PERSISTENT context throughout the track — do NOT lose state.
+
+## Startup
+
+1. Read `.claude/commands/mma-tier2-tech-lead.md` — load your role definition and hard rules FIRST
+2. Read `conductor/workflow.md` for the full task lifecycle protocol
+3. Read `conductor/tech-stack.md` for technology constraints
+4. Read the target track's `spec.md` and `plan.md`
+5. Identify the current task: first `[ ]` or `[~]` in `plan.md`
+
+If no track name is provided, run `/conductor-status` first and ask which track to implement.
+
+## Task Lifecycle (per task)
+
+Follow this EXACTLY per `conductor/workflow.md`:
+
+### 1. Mark In Progress
+Edit `plan.md`: change `[ ]` → `[~]` for the current task.
+
+### 2. Research Phase (High-Signal)
+Before touching code, use context-efficient tools IN THIS ORDER:
+1. `py_get_code_outline` — FIRST call on any Python file. Maps functions/classes with line ranges.
+2. `py_get_skeleton` — signatures + docstrings only, no bodies
+3. `get_git_diff` — understand recent changes before modifying touched files
+4. `Grep`/`Glob` — cross-file symbol search
+5. `Read` (targeted, offset+limit only) — ONLY after outline identifies specific ranges
+
+**NEVER** call `Read` on a full Python file >50 lines without a prior `py_get_code_outline` call.
+
+### 3. Write Failing Tests (Red Phase — TDD)
+**DELEGATE to Tier 3 Worker** — do NOT write tests yourself:
+```powershell
+uv run python scripts\claude_mma_exec.py --role tier3-worker "Write failing tests for: {TASK_DESCRIPTION}. Focus files: {FILE_LIST}. Spec: {RELEVANT_SPEC_EXCERPT}"
+```
+Run the tests. Confirm they FAIL. This is the Red phase.
+
+### 4. Implement to Pass (Green Phase)
+**DELEGATE to Tier 3 Worker**:
+```powershell
+uv run python scripts\claude_mma_exec.py --role tier3-worker "Implement minimum code to pass these tests: {TEST_FILE}. Focus files: {FILE_LIST}"
+```
+Run tests. Confirm they PASS. This is the Green phase.
+
+### 5. Refactor (Optional)
+With passing tests as safety net, refactor if needed. Rerun tests.
+
+### 6. Verify Coverage
+Use `run_powershell` MCP tool (not Bash — Bash is a mingw sandbox on Windows):
+```powershell
+uv run pytest --cov=. --cov-report=term-missing {TEST_FILE}
+```
+Target: >80% for new code.
+
+### 7. Commit
+Stage changes. Message format:
+```
+feat({scope}): {description}
+```
+
+### 8. Attach Git Notes
+```powershell
+$sha = git log -1 --format="%H"
+git notes add -m "Task: {TASK_NAME}`nSummary: {CHANGES}`nFiles: {FILE_LIST}" $sha
+```
+
+### 9. Update plan.md
+Change `[~]` → `[x]` and append first 7 chars of commit SHA:
+```
+[x] Task description. abc1234
+```
+Commit: `conductor(plan): Mark task '{TASK_NAME}' as complete`
+
+### 10. Next Task or Phase Completion
+- If more tasks in current phase: loop to step 1 with next task
+- If phase complete: run `/conductor-verify`
+
+## Error Handling
+
+### Tier 3 delegation fails (credit limit, API error, timeout)
+**STOP** — do NOT implement inline as a fallback. Ask the user:
+> "Tier 3 Worker is unavailable ({reason}). Should I continue with a different provider, or wait?"
+Never silently absorb Tier 3 work into Tier 2 context.
+
+### Tests fail with large output — delegate to Tier 4 QA:
+```powershell
+uv run python scripts\claude_mma_exec.py --role tier4-qa "Analyze this test failure: {ERROR_SUMMARY}. Test file: {TEST_FILE}"
+```
+Maximum 2 fix attempts. If still failing: STOP and ask the user.
+
+## Deviations from Tech Stack
+If implementation requires something not in `tech-stack.md`:
+1. **STOP** implementation
+2. Update `tech-stack.md` with justification
+3. Add dated note
+4. Resume
+
+## Important
+- You are Tier 2 — delegate heavy implementation to Tier 3
+- Maintain persistent context across the entire track
+- Use Research-First Protocol before reading large files
+- The plan.md is the SOURCE OF TRUTH for task state

.claude/commands/conductor-new-track.md

@@ -0,0 +1,174 @@
+---
+description: Initialize a new conductor track with spec, plan, and metadata
+---
+
+# /conductor-new-track
+
+Create a new track in the conductor system. This is a Tier 1 (Orchestrator) operation.
+The quality of the spec and plan directly determines whether Tier 3 workers can execute
+without confusion. Vague specs produce vague implementations.
+
+## Prerequisites
+- Read `conductor/product.md` and `conductor/product-guidelines.md` for product alignment
+- Read `conductor/tech-stack.md` for technology constraints
+- Consult architecture docs in `docs/` when the track touches core systems:
+  - `docs/guide_architecture.md`: Threading, events, AI client, HITL mechanism
+  - `docs/guide_tools.md`: MCP tools, Hook API, ApiHookClient
+  - `docs/guide_mma.md`: Tickets, tracks, DAG engine, worker lifecycle
+  - `docs/guide_simulations.md`: Test framework, mock provider, verification patterns
+
+## Steps
+
+### 1. Gather Information
+Ask the user for:
+- **Track name**: descriptive, snake_case (e.g., `add_auth_system`)
+- **Track type**: `feat`, `fix`, `refactor`, `chore`
+- **Description**: one-line summary
+- **Requirements**: functional requirements for the spec
+
+### 2. MANDATORY: Deep Codebase Audit
+
+**This step is what separates useful specs from useless ones.**
+
+Before writing a single line of spec, you MUST audit the actual codebase to understand
+what already exists. Use the Research-First Protocol:
+
+1. **Map the target area**: Use `py_get_code_outline` on every file the track will touch.
+   Identify existing functions, classes, and their line ranges.
+2. **Read key implementations**: Use `py_get_definition` on functions that are relevant
+   to the track's goals. Understand their signatures, data structures, and control flow.
+3. **Search for existing work**: Use `Grep` to find symbols, patterns, or partial
+   implementations that may already address some requirements.
+4. **Check recent changes**: Use `get_git_diff` on target files to understand what's
+   been modified recently and by which tracks.
+
+**Output of this step**: A "Current State Audit" section listing:
+- What already exists (with file:line references)
+- What's missing (the actual gaps this track fills)
+- What's partially implemented and needs enhancement
+
+### 3. Create Track Directory
+```
+conductor/tracks/{track_name}_{YYYYMMDD}/
+```
+Use today's date in YYYYMMDD format.
+
+### 4. Create metadata.json
+```json
+{
+  "track_id": "{track_name}_{YYYYMMDD}",
+  "type": "{feat|fix|refactor|chore}",
+  "status": "new",
+  "created_at": "{ISO8601}",
+  "updated_at": "{ISO8601}",
+  "description": "{description}"
+}
+```
+
+### 5. Create index.md
+```markdown
+# Track {track_name}_{YYYYMMDD} Context
+
+- [Specification](./spec.md)
+- [Implementation Plan](./plan.md)
+- [Metadata](./metadata.json)
+```
+
+### 6. Create spec.md — The Surgical Specification
+
+The spec MUST include these sections:
+
+```markdown
+# Track Specification: {Title}
+
+## Overview
+{What this track delivers and WHY — 2-3 sentences max}
+
+## Current State Audit (as of {latest_commit_sha})
+### Already Implemented (DO NOT re-implement)
+- **{Feature}** (`{function_name}`, {file}:{lines}): {what it does}
+- ...
+
+### Gaps to Fill (This Track's Scope)
+1. **{Gap}**: {What's missing, with reference to where it should go}
+2. ...
+
+## Goals
+{Numbered list — crisp, no fluff}
+
+## Functional Requirements
+### {Requirement Group}
+- {Specific requirement referencing actual data structures, function names, dict keys}
+- ...
+
+## Non-Functional Requirements
+- Thread safety constraints (reference guide_architecture.md if applicable)
+- Performance targets
+- No new dependencies unless justified
+
+## Architecture Reference
+- {Link to relevant docs/guide_*.md section}
+
+## Out of Scope
+- {Explicit exclusions}
+```
+
+**Critical rules for specs:**
+- NEVER describe a feature to implement without first checking if it exists
+- ALWAYS include the "Current State Audit" section with line references
+- ALWAYS link to relevant architecture docs
+- Reference actual variable names, dict keys, and class names from the codebase
+
+### 7. Create plan.md — The Surgical Plan
+
+Each task must be specific enough that a Tier 3 worker on a lightweight model
+can execute it without needing to understand the overall architecture.
+
+```markdown
+# Implementation Plan: {Title}
+
+Architecture reference: [docs/guide_architecture.md](../../docs/guide_architecture.md)
+
+## Phase 1: {Phase Name}
+Focus: {One-sentence scope}
+
+- [ ] Task 1.1: {SURGICAL description — see rules below}
+- [ ] Task 1.2: ...
+- [ ] Task 1.N: Write tests for {what Phase 1 changed}
+- [ ] Task 1.X: Conductor - User Manual Verification (Protocol in workflow.md)
+```
+
+**Rules for writing tasks:**
+
+1. **Reference exact locations**: "In `_render_mma_dashboard` (gui_2.py:2700-2701)"
+   not "in the dashboard."
+2. **Specify the API**: "Use `imgui.progress_bar(value, ImVec2(-1, 0), label)`"
+   not "add a progress bar."
+3. **Name the data**: "Read from `self.mma_streams` dict, keys prefixed with `'Tier 3'`"
+   not "display the streams."
+4. **Describe the change shape**: "Replace the single text box with four collapsible sections"
+   not "improve the display."
+5. **State thread safety**: "Push via `_pending_gui_tasks` with lock" when the task
+   involves cross-thread data.
+6. **For bug fixes**: List specific root cause candidates with code-level reasoning,
+   not "investigate and fix."
+7. **Each phase ends with**: A test task and a verification task.
+
+### 8. Commit
+```
+conductor(track): Initialize track '{track_name}'
+```
+
+## Anti-Patterns (DO NOT do these)
+
+- **Spec that describes features without checking if they exist** → produces duplicate work
+- **Task that says "implement X" without saying WHERE or HOW** → worker guesses wrong
+- **Plan with no line references** → worker wastes tokens searching
+- **Spec with no architecture doc links** → worker misunderstands threading/data model
+- **Tasks scoped too broadly** → worker tries to do too much, fails
+- **No "Current State Audit"** → entire track may be re-implementing existing code
+
+## Important
+- Do NOT start implementing — track initialization only
+- Implementation is done via `/conductor-implement`
+- Each task should be scoped for a single Tier 3 Worker delegation

.claude/commands/conductor-setup.md

@@ -0,0 +1,46 @@
+---
+description: Initialize conductor context — read product docs, verify structure, report readiness
+---
+
+# /conductor-setup
+
+Bootstrap a Claude Code session with full conductor context. Run this at session start.
+
+## Steps
+
+1. **Read Core Documents:**
+   - `conductor/index.md` — navigation hub
+   - `conductor/product.md` — product vision
+   - `conductor/product-guidelines.md` — UX/code standards
+   - `conductor/tech-stack.md` — technology constraints
+   - `conductor/workflow.md` — task lifecycle (skim; reference during implementation)
+
+2. **Check Active Tracks:**
+   - List all directories in `conductor/tracks/`
+   - Read each `metadata.json` for status
+   - Read each `plan.md` for current task state
+   - Identify the track with `[~]` in-progress tasks
+
+3. **Check Session Context:**
+   - Read `TASKS.md` if it exists — check for IN_PROGRESS or BLOCKED tasks
+   - Read last 3 entries in `JOURNAL.md` for recent activity
+   - Run `git log --oneline -10` for recent commits
+
+4. **Report Readiness:**
+   Present a session startup summary:
+   ```
+   ## Session Ready
+
+   **Active Track:** {track name} — Phase {N}, Task: {current task description}
+   **Recent Activity:** {last journal entry title}
+   **Last Commit:** {git log -1 oneline}
+
+   Ready to:
+   - `/conductor-implement` — resume active track
+   - `/conductor-status` — full status overview
+   - `/conductor-new-track` — start new work
+   ```
+
+## Important
+- This is READ-ONLY — do not modify files
+- This replaces Gemini's `activate_skill mma-orchestrator` + `/conductor:setup`

.claude/commands/conductor-status.md

@@ -0,0 +1,32 @@
+---
+description: Show current conductor track status — active tracks, phases, pending tasks
+---
+
+# /conductor-status
+
+Read the conductor track registry and all active tracks, then report current project state.
+
+## Steps
+
+1. Read `conductor/tracks.md` for the track registry
+2. For each track directory in `conductor/tracks/`:
+   - Read `metadata.json` for status
+   - Read `plan.md` and count: total tasks, completed `[x]`, in-progress `[~]`, pending `[ ]`
+   - Identify the current phase (first phase with `[~]` or `[ ]` tasks)
+3. Read `JOURNAL.md` last 3 entries for recent activity context
+
+## Output Format
+
+Present a summary table:
+
+```
+| Track | Status | Phase | Progress | Last SHA |
+|-------|--------|-------|----------|----------|
+```
+
+Then for each in-progress track, list the specific next pending task.
+
+## Important
+- This is READ-ONLY — do not modify any files
+- Report exactly what the plan.md files say
+- Flag any discrepancies (e.g., metadata says "new" but plan.md has [x] tasks)

.claude/commands/conductor-verify.md

@@ -0,0 +1,85 @@
+---
+description: Run phase completion verification — tests, coverage, checkpoint commit
+---
+
+# /conductor-verify
+
+Execute the Phase Completion Verification and Checkpointing Protocol.
+Run this when all tasks in a phase are marked `[x]`.
+
+## Protocol
+
+### 1. Announce
+Tell the user: "Phase complete. Running verification and checkpointing protocol."
+
+### 2. Verify Test Coverage for Phase
+
+Find the phase scope:
+- Read `plan.md` to find the previous phase's checkpoint SHA
+- If no previous checkpoint: scope is all changes since first commit
+- Run: `git diff --name-only {previous_checkpoint_sha} HEAD`
+- For each changed code file (exclude `.json`, `.md`, `.yaml`, `.toml`):
+  - Check if a corresponding test file exists
+  - If missing: create one (analyze existing test style first)
+
+### 3. Run Automated Tests
+
+**ANNOUNCE the exact command before running:**
+> "I will now run the automated test suite. Command: `uv run pytest --cov=. --cov-report=term-missing -x`"
+
+Execute the command.
+
+**If tests fail with large output:**
+- Pipe output to `logs/phase_verify.log`
+- Spawn Tier 4 QA for analysis:
+```powershell
+uv run python scripts\claude_mma_exec.py --role tier4-qa "Analyze test failures from logs/phase_verify.log"
+```
+- Maximum 2 fix attempts
+- If still failing: **STOP**, report to user, await guidance
+
+### 4. API Hook Verification (if applicable)
+
+If the track involves UI changes:
+- Check if GUI test hooks are available on port 8999
+- Run relevant simulation tests from `tests/visual_sim_*.py`
+- Log results
+
+### 5. Present Results and WAIT
+
+Display:
+- Test results (pass/fail count)
+- Coverage report
+- Any verification logs
+
+**PAUSE HERE.** Do NOT proceed without explicit user confirmation.
+
+### 6. Create Checkpoint Commit
+
+After user confirms:
+```powershell
+git add -A
+git commit -m "conductor(checkpoint): Checkpoint end of Phase {N} - {Phase Name}"
+```
+
+### 7. Attach Verification Report via Git Notes
+```powershell
+$sha = git log -1 --format="%H"
+git notes add -m "Phase Verification Report`nCommand: {test_command}`nResult: {pass/fail}`nCoverage: {percentage}`nConfirmed by: user" $sha
+```
+
+### 8. Update plan.md
+
+Update the phase heading to include checkpoint SHA:
+```markdown
+## Phase N: {Name} [checkpoint: {sha_7}]
+```
+Commit: `conductor(plan): Mark phase '{Phase Name}' as complete`
+
+### 9. Announce Completion
+Tell the user the phase is complete with a summary of the verification report.
+
+## Context Reset
+After phase checkpointing, treat the checkpoint as ground truth.
+Prior conversational context about implementation details can be dropped.
+The checkpoint commit and git notes preserve the audit trail.

.claude/commands/mma-tier1-orchestrator.md

@@ -0,0 +1,72 @@
+---
+description: Tier 1 Orchestrator — product alignment, high-level planning, track initialization
+---
+
+STRICT SYSTEM DIRECTIVE: You are a Tier 1 Orchestrator. Focused on product alignment, high-level planning, and track initialization. ONLY output the requested text. No pleasantries.
+
+# MMA Tier 1: Orchestrator
+
+## Primary Context Documents
+Read at session start: `conductor/product.md`, `conductor/product-guidelines.md`
+
+## Architecture Fallback
+When planning tracks that touch core systems, consult the deep-dive docs:
+- `docs/guide_architecture.md`: Thread domains, event system, AI client, HITL mechanism, frame-sync action catalog
+- `docs/guide_tools.md`: MCP Bridge security, 26-tool inventory, Hook API endpoints, ApiHookClient
+- `docs/guide_mma.md`: Ticket/Track data structures, DAG engine, ConductorEngine, worker lifecycle
+- `docs/guide_simulations.md`: live_gui fixture, Puppeteer pattern, mock provider, verification patterns
+
+## Responsibilities
+- Maintain alignment with the product guidelines and definition
+- Define track boundaries and initialize new tracks (`/conductor-new-track`)
+- Set up the project environment (`/conductor-setup`)
+- Delegate track execution to the Tier 2 Tech Lead
+
+## The Surgical Methodology
+
+When creating or refining tracks, follow this protocol to produce specs that
+lesser-reasoning models can execute without confusion:
+
+### 1. Audit Before Specifying
+NEVER write a spec without first reading the actual code. Use `py_get_code_outline`,
+`py_get_definition`, `Grep`, and `get_git_diff` to build a map of what exists.
+Document existing implementations with file:line references in a "Current State Audit"
+section. This prevents specs that ask to re-implement existing features.
+
+### 2. Identify Gaps, Not Features
+The spec should focus on what's MISSING, not what the track "will build."
+Frame requirements as: "The existing `_render_mma_dashboard` (gui_2.py:2633-2724)
+has a token usage table but no cost estimation column. Add cost tracking."
+Not: "Build a metrics dashboard with token and cost tracking."
+
+### 3. Write Worker-Ready Tasks
+Each task in the plan must be executable by a Tier 3 worker on a lightweight model
+(gemini-2.5-flash-lite) without needing to understand the overall architecture.
+This means every task must specify:
+- **WHERE**: Exact file and line range to modify
+- **WHAT**: The specific change (add function, modify dict, extend table)
+- **HOW**: Which API calls, data structures, or patterns to use
+- **SAFETY**: Thread-safety constraints if cross-thread data is involved
+
+### 4. Reference Architecture Docs
+Every spec should link to the relevant `docs/guide_*.md` section so implementing
+agents have a fallback when confused about threading, data flow, or module interactions.
+
+### 5. Map Dependencies
+Explicitly state which tracks must complete before this one, and which tracks
+this one blocks. Include execution order in the spec.
+
+### 6. Root Cause Analysis (for fix tracks)
+Don't write "investigate and fix X." Instead, read the code, trace the data flow,
+and list specific root cause candidates with code-level reasoning:
+"Candidate 1: `_queue_put` (line 138) uses `asyncio.run_coroutine_threadsafe` but
+the `else` branch uses `put_nowait` which is NOT thread-safe from a thread-pool thread."
+
+## Limitations
+- Read-only tools only: Read, Glob, Grep, WebFetch, WebSearch, Bash (read-only ops)
+- Do NOT execute tracks or implement features
+- Do NOT write code or edit files (except track spec/plan/metadata)
+- Do NOT perform low-level bug fixing
+- Keep context strictly focused on product definitions and high-level strategy
+- To delegate track execution: instruct the human operator to run:
+  `uv run python scripts\claude_mma_exec.py --role tier2-tech-lead "[PROMPT]"`

.claude/commands/mma-tier2-tech-lead.md

@@ -0,0 +1,74 @@
+---
+description: Tier 2 Tech Lead — track execution, architectural oversight, delegation to Tier 3/4
+---
+
+STRICT SYSTEM DIRECTIVE: You are a Tier 2 Tech Lead. Focused on architectural design and track execution. ONLY output the requested text. No pleasantries.
+
+# MMA Tier 2: Tech Lead
+
+## Primary Context Documents
+Read at session start: `conductor/tech-stack.md`, `conductor/workflow.md`
+
+## Responsibilities
+- Manage the execution of implementation tracks (`/conductor-implement`)
+- Ensure alignment with `tech-stack.md` and project architecture
+- Break down tasks into specific technical steps for Tier 3 Workers
+- Maintain PERSISTENT context throughout a track's implementation phase (NO Context Amnesia)
+- Review implementations and coordinate bug fixes via Tier 4 QA
+- **CRITICAL: ATOMIC PER-TASK COMMITS**: You MUST commit your progress on a per-task basis. Immediately after a task is verified successfully, you must stage the changes, commit them, attach the git note summary, and update `plan.md` before moving to the next task. Do NOT batch multiple tasks into a single commit.
+- **Meta-Level Sanity Check**: After completing a track (or upon explicit request), perform a codebase sanity check. Run `uv run ruff check .` and `uv run mypy --explicit-package-bases .` to ensure Tier 3 Workers haven't degraded static analysis constraints. Identify broken simulation tests and append them to a tech debt track or fix them immediately.
+
+## Delegation Commands (PowerShell)
+
+```powershell
+# Spawn Tier 3 Worker for implementation tasks
+uv run python scripts\claude_mma_exec.py --role tier3-worker "[PROMPT]"
+
+# Spawn Tier 4 QA Agent for error analysis
+uv run python scripts\claude_mma_exec.py --role tier4-qa "[PROMPT]"
+```
+
+### @file Syntax for Tier 3 Context Injection
+`@filepath` anywhere in the prompt string is detected by `claude_mma_exec.py` and the file is automatically inlined into the Tier 3 context. Use this so Tier 3 has what it needs WITHOUT Tier 2 reading those files first.
+
+```powershell
+# Example: Tier 3 gets api_hook_client.py and the styleguide injected automatically
+uv run python scripts\claude_mma_exec.py --role tier3-worker "Apply type hints to @api_hook_client.py following @conductor/code_styleguides/python.md. ..."
+```
+
+## Tool Use Hierarchy (MANDATORY — enforced order)
+
+Claude has access to all tools and will default to familiar ones. This hierarchy OVERRIDES that default.
+
+**For any Python file investigation, use in this order:**
+1. `py_get_code_outline` — structure map (functions, classes, line ranges). Use this FIRST.
+2. `py_get_skeleton` — signatures + docstrings, no bodies
+3. `get_file_summary` — high-level prose summary
+4. `py_get_definition` / `py_get_signature` — targeted symbol lookup
+5. `Grep` / `Glob` — cross-file symbol search and pattern matching
+6. `Read` (targeted, with offset/limit) — ONLY after outline identifies specific line ranges
+
+**`run_powershell` (MCP tool)** — PRIMARY shell execution on Windows. Use for: git, tests, scan scripts, any shell command. This is native PowerShell, not bash/mingw.
+
+**Bash** — LAST RESORT only when MCP server is not running. Bash runs in a mingw sandbox on Windows and may produce no output. Prefer `run_powershell` for everything.
+
+## Hard Rules (Non-Negotiable)
+
+- **NEVER** call `Read` on a file >50 lines without calling `py_get_code_outline` or `py_get_skeleton` first.
+- **NEVER** write implementation code, refactor code, type hint code, or test code inline in this context. If it goes into the codebase, Tier 3 writes it.
+- **NEVER** write or run inline Python scripts via Bash. If a script is needed, it already exists or Tier 3 creates it.
+- **NEVER** process raw bash output for large outputs inline — write to a file and Read, or delegate to Tier 4 QA.
+- **ALWAYS** use `@file` injection in Tier 3 prompts rather than reading and summarizing files yourself.
+
+## Refactor-Heavy Tracks (Type Hints, Style Sweeps)
+
+For tracks with no new logic — only mechanical code changes (type hints, style fixes, renames):
+- **No TDD cycle required.** Skip Red/Green phases. The verification is: scan report shows 0 remaining items.
+- Tier 2 role: scope the batch, write a precise Tier 3 prompt, delegate, verify with scan script.
+- Batch by file group. One Tier 3 call per group (e.g., all scripts/, all simulation/).
+- Verification command: `uv run python scripts\scan_all_hints.py` then read `scan_report.txt`
+
+## Limitations
+- Do NOT perform heavy implementation work directly — delegate to Tier 3
+- Do NOT write test or implementation code directly
+- For large error logs, always spawn Tier 4 QA rather than reading raw stderr

.claude/commands/mma-tier3-worker.md

@@ -0,0 +1,22 @@
+---
+description: Tier 3 Worker — stateless TDD implementation, surgical code changes
+---
+
+STRICT SYSTEM DIRECTIVE: You are a stateless Tier 3 Worker (Contributor). Your goal is to implement specific code changes or tests based on the provided task. You have access to tools for reading and writing files (Read, Write, Edit), codebase investigation (Glob, Grep), version control (Bash git commands), and web tools (WebFetch, WebSearch). You CAN execute PowerShell scripts via Bash for verification and testing. Follow TDD and return success status or code changes. No pleasantries, no conversational filler.
+
+# MMA Tier 3: Worker
+
+## Context Model: Context Amnesia
+Treat each invocation as starting from zero. Use ONLY what is provided in this prompt plus files you explicitly read during this session. Do not reference prior conversation history.
+
+## Responsibilities
+- Implement code strictly according to the provided prompt and specifications
+- Write failing tests FIRST (Red phase), then implement code to pass them (Green phase)
+- Ensure all changes are minimal, surgical, and conform to the requested standards
+- Utilize tool access (Read, Write, Edit, Glob, Grep, Bash) to implement and verify
+
+## Limitations
+- No architectural decisions — if ambiguous, pick the minimal correct approach and note the assumption
+- No modifications to unrelated files beyond the immediate task scope
+- Stateless — always assume a fresh context per invocation
+- Rely on dependency skeletons provided in the prompt for understanding module interfaces

.claude/commands/mma-tier4-qa.md

@@ -0,0 +1,30 @@
+---
+description: Tier 4 QA Agent — stateless error analysis, log summarization, no fixes
+---
+
+STRICT SYSTEM DIRECTIVE: You are a stateless Tier 4 QA Agent. Your goal is to analyze errors, summarize logs, or verify tests. Read-only access only. Do NOT implement fixes. Do NOT modify any files. ONLY output the requested analysis. No pleasantries.
+
+# MMA Tier 4: QA Agent
+
+## Context Model: Context Amnesia
+Stateless — treat each invocation as a fresh context. Use only what is provided in this prompt and files you explicitly read.
+
+## Responsibilities
+- Compress large stack traces or log files into concise, actionable summaries
+- Identify the root cause of test failures or runtime errors
+- Provide a brief, technical description of the required fix (description only — NOT the implementation)
+- Utilize diagnostic tools (Read, Glob, Grep, Bash read-only) to verify failures
+
+## Output Format
+
+```
+ROOT CAUSE: [one sentence]
+AFFECTED FILE: [path:line if identifiable]
+RECOMMENDED FIX: [one sentence description for Tier 2 to action]
+```
+
+## Limitations
+- Do NOT implement the fix directly
+- Do NOT write or modify any files
+- Ensure output is extremely brief and focused
+- Always operate statelessly — assume fresh context each invocation

.claude/settings.json

@@ -0,0 +1,3 @@
+{
+  "outputStyle": "default"
+}

.claude/settings.local.json

@@ -0,0 +1,22 @@
+{
+  "permissions": {
+    "allow": [
+      "mcp__manual-slop__run_powershell",
+      "mcp__manual-slop__py_get_definition",
+      "mcp__manual-slop__read_file",
+      "mcp__manual-slop__py_get_code_outline",
+      "mcp__manual-slop__get_file_slice",
+      "mcp__manual-slop__py_find_usages",
+      "mcp__manual-slop__set_file_slice",
+      "mcp__manual-slop__py_check_syntax",
+      "mcp__manual-slop__get_file_summary",
+      "mcp__manual-slop__get_tree",
+      "mcp__manual-slop__list_directory",
+      "mcp__manual-slop__py_get_skeleton"
+    ]
+  },
+  "enableAllProjectMcpServers": true,
+  "enabledMcpjsonServers": [
+    "manual-slop"
+  ]
+}

.dockerignore

@@ -0,0 +1,21 @@
+.venv
+__pycache__
+*.pyc
+*.pyo
+*.pyd
+.git
+.gitignore
+logs
+gallery
+md_gen
+credentials.toml
+manual_slop.toml
+manual_slop_history.toml
+manualslop_layout.ini
+dpg_layout.ini
+.pytest_cache
+scripts/generated
+.gemini
+conductor/archive
+.editorconfig
+*.log

.editorconfig

@@ -2,7 +2,7 @@ root = true
 
 [*.py]
 indent_style = space
-indent_size  = 2
+indent_size  = 1
 
 [*.s]
 indent_style = tab

.gemini/agents/tier1-orchestrator.md

@@ -0,0 +1,100 @@
+---
+name: tier1-orchestrator
+description: Tier 1 Orchestrator for product alignment and high-level planning.
+model: gemini-3.1-pro-preview
+tools:
+  - read_file
+  - list_directory
+  - discovered_tool_search_files
+  - grep_search
+  - discovered_tool_get_file_summary
+  - discovered_tool_get_python_skeleton
+  - discovered_tool_get_code_outline
+  - discovered_tool_get_git_diff
+  - discovered_tool_web_search
+  - discovered_tool_fetch_url
+  - activate_skill
+  - discovered_tool_run_powershell
+  - discovered_tool_py_find_usages
+  - discovered_tool_py_get_imports
+  - discovered_tool_py_check_syntax
+  - discovered_tool_py_get_hierarchy
+  - discovered_tool_py_get_docstring
+  - discovered_tool_get_tree
+  - discovered_tool_py_get_definition
+---
+STRICT SYSTEM DIRECTIVE: You are a Tier 1 Orchestrator.
+Focused on product alignment, high-level planning, and track initialization.
+ONLY output the requested text. No pleasantries.
+
+## Architecture Fallback
+When planning tracks that touch core systems, consult the deep-dive docs:
+- `docs/guide_architecture.md`: Thread domains, event system, AI client, HITL mechanism, frame-sync action catalog
+- `docs/guide_tools.md`: MCP Bridge security, 26-tool inventory, Hook API endpoints, ApiHookClient
+- `docs/guide_mma.md`: Ticket/Track data structures, DAG engine, ConductorEngine, worker lifecycle
+- `docs/guide_simulations.md`: live_gui fixture, Puppeteer pattern, mock provider, verification patterns
+
+## The Surgical Methodology
+
+When creating or refining tracks, you MUST follow this protocol:
+
+### 1. MANDATORY: Audit Before Specifying
+NEVER write a spec without first reading the actual code using your tools.
+Use `get_code_outline`, `py_get_definition`, `grep_search`, and `get_git_diff`
+to build a map of what exists. Document existing implementations with file:line
+references in a "Current State Audit" section in the spec.
+
+**WHY**: Previous track specs asked to implement features that already existed
+(Track Browser, DAG tree, approval dialogs) because no code audit was done first.
+This wastes entire implementation phases.
+
+### 2. Identify Gaps, Not Features
+Frame requirements around what's MISSING relative to what exists:
+GOOD: "The existing `_render_mma_dashboard` (gui_2.py:2633-2724) has a token
+usage table but no cost estimation column."
+BAD: "Build a metrics dashboard with token and cost tracking."
+
+### 3. Write Worker-Ready Tasks
+Each plan task must be executable by a Tier 3 worker on gemini-2.5-flash-lite
+without understanding the overall architecture. Every task specifies:
+- **WHERE**: Exact file and line range (`gui_2.py:2700-2701`)
+- **WHAT**: The specific change (add function, modify dict, extend table)
+- **HOW**: Which API calls or patterns (`imgui.progress_bar(...)`, `imgui.collapsing_header(...)`)
+- **SAFETY**: Thread-safety constraints if cross-thread data is involved
+
+### 4. For Bug Fix Tracks: Root Cause Analysis
+Don't write "investigate and fix." Read the code, trace the data flow, list
+specific root cause candidates with code-level reasoning.
+
+### 5. Reference Architecture Docs
+Link to relevant `docs/guide_*.md` sections in every spec so implementing
+agents have a fallback for threading, data flow, or module interactions.
+
+### 6. Map Dependencies Between Tracks
+State execution order and blockers explicitly in metadata.json and spec.
+
+## Spec Template (REQUIRED sections)
+```
+# Track Specification: {Title}
+
+## Overview
+## Current State Audit (as of {commit_sha})
+### Already Implemented (DO NOT re-implement)
+### Gaps to Fill (This Track's Scope)
+## Goals
+## Functional Requirements
+## Non-Functional Requirements
+## Architecture Reference
+## Out of Scope
+```
+
+## Plan Template (REQUIRED format)
+```
+## Phase N: {Name}
+Focus: {One-sentence scope}
+
+- [ ] Task N.1: {Surgical description with file:line refs and API calls}
+- [ ] Task N.2: ...
+- [ ] Task N.N: Write tests for Phase N changes
+- [ ] Task N.X: Conductor - User Manual Verification (Protocol in workflow.md)
+```

.gemini/agents/tier2-tech-lead.md

@@ -0,0 +1,29 @@
+---
+name: tier2-tech-lead
+description: Tier 2 Tech Lead for architectural design and execution.
+model: gemini-3-flash-preview
+tools:
+  - read_file
+  - write_file
+  - replace
+  - list_directory
+  - discovered_tool_search_files
+  - grep_search
+  - discovered_tool_get_file_summary
+  - discovered_tool_get_python_skeleton
+  - discovered_tool_get_code_outline
+  - discovered_tool_get_git_diff
+  - discovered_tool_web_search
+  - discovered_tool_fetch_url
+  - activate_skill
+  - discovered_tool_run_powershell
+  - discovered_tool_py_find_usages
+  - discovered_tool_py_get_imports
+  - discovered_tool_py_check_syntax
+  - discovered_tool_py_get_hierarchy
+  - discovered_tool_py_get_docstring
+  - discovered_tool_get_tree
+---
+STRICT SYSTEM DIRECTIVE: You are a Tier 2 Tech Lead.
+Focused on architectural design and track execution.
+ONLY output the requested text. No pleasantries.

.gemini/agents/tier3-worker.md

@@ -0,0 +1,31 @@
+---
+name: tier3-worker
+description: Stateless Tier 3 Worker for code implementation and TDD.
+model: gemini-3-flash-preview
+tools:
+  - read_file
+  - write_file
+  - replace
+  - list_directory
+  - discovered_tool_search_files
+  - grep_search
+  - discovered_tool_get_file_summary
+  - discovered_tool_get_python_skeleton
+  - discovered_tool_get_code_outline
+  - discovered_tool_get_git_diff
+  - discovered_tool_web_search
+  - discovered_tool_fetch_url
+  - activate_skill
+  - discovered_tool_run_powershell
+  - discovered_tool_py_find_usages
+  - discovered_tool_py_get_imports
+  - discovered_tool_py_check_syntax
+  - discovered_tool_py_get_hierarchy
+  - discovered_tool_py_get_docstring
+  - discovered_tool_get_tree
+---
+STRICT SYSTEM DIRECTIVE: You are a stateless Tier 3 Worker (Contributor).
+Your goal is to implement specific code changes or tests based on the provided task.
+You have access to tools for reading and writing files, codebase investigation, and web tools.
+You CAN execute PowerShell scripts or run shell commands via discovered_tool_run_powershell for verification and testing.
+Follow TDD and return success status or code changes. No pleasantries, no conversational filler.

.gemini/agents/tier4-qa.md

@@ -0,0 +1,29 @@
+---
+name: tier4-qa
+description: Stateless Tier 4 QA Agent for log analysis and diagnostics.
+model: gemini-2.5-flash-lite
+tools:
+  - read_file
+  - list_directory
+  - discovered_tool_search_files
+  - grep_search
+  - discovered_tool_get_file_summary
+  - discovered_tool_get_python_skeleton
+  - discovered_tool_get_code_outline
+  - discovered_tool_get_git_diff
+  - discovered_tool_web_search
+  - discovered_tool_fetch_url
+  - activate_skill
+  - discovered_tool_run_powershell
+  - discovered_tool_py_find_usages
+  - discovered_tool_py_get_imports
+  - discovered_tool_py_check_syntax
+  - discovered_tool_py_get_hierarchy
+  - discovered_tool_py_get_docstring
+  - discovered_tool_get_tree
+---
+STRICT SYSTEM DIRECTIVE: You are a stateless Tier 4 QA Agent.
+Your goal is to analyze errors, summarize logs, or verify tests.
+You have access to tools for reading files, exploring the codebase, and web tools.
+You CAN execute PowerShell scripts or run shell commands via discovered_tool_run_powershell for diagnostics.
+ONLY output the requested analysis. No pleasantries.

.gemini/policies/99-agent-full-autonomy.toml

@@ -0,0 +1,269 @@
+[[rule]] 
+toolName = "discovered_tool_fetch_url"
+decision = "allow"
+priority = 100
+description = "Allow discovered fetch_url tool."
+
+[[rule]]
+toolName = "discovered_tool_get_file_slice"
+decision = "allow"
+priority = 100
+description = "Allow discovered get_file_slice tool."
+
+[[rule]]
+toolName = "discovered_tool_get_file_summary"
+decision = "allow"
+priority = 100
+description = "Allow discovered get_file_summary tool."
+
+[[rule]]
+toolName = "discovered_tool_get_git_diff"
+decision = "allow"
+priority = 100
+description = "Allow discovered get_git_diff tool."
+
+[[rule]]
+toolName = "discovered_tool_get_tree"
+decision = "allow"
+priority = 100
+description = "Allow discovered get_tree tool."
+
+[[rule]]
+toolName = "discovered_tool_get_ui_performance"
+decision = "allow"
+priority = 100
+description = "Allow discovered get_ui_performance tool."
+
+[[rule]]
+toolName = "discovered_tool_list_directory"
+decision = "allow"
+priority = 100
+description = "Allow discovered list_directory tool."
+
+[[rule]]
+toolName = "discovered_tool_py_check_syntax"
+decision = "allow"
+priority = 100
+description = "Allow discovered py_check_syntax tool."
+
+[[rule]]
+toolName = "discovered_tool_py_find_usages"
+decision = "allow"
+priority = 100
+description = "Allow discovered py_find_usages tool."
+
+[[rule]]
+toolName = "discovered_tool_py_get_class_summary"
+decision = "allow"
+priority = 100
+description = "Allow discovered py_get_class_summary tool."
+
+[[rule]]
+toolName = "discovered_tool_py_get_code_outline"
+decision = "allow"
+priority = 100
+description = "Allow discovered py_get_code_outline tool."
+
+[[rule]]
+toolName = "discovered_tool_py_get_definition"
+decision = "allow"
+priority = 100
+description = "Allow discovered py_get_definition tool."
+
+[[rule]]
+toolName = "discovered_tool_py_get_docstring"
+decision = "allow"
+priority = 100
+description = "Allow discovered py_get_docstring tool."
+
+[[rule]]
+toolName = "discovered_tool_py_get_hierarchy"
+decision = "allow"
+priority = 100
+description = "Allow discovered py_get_hierarchy tool."
+
+[[rule]]
+toolName = "discovered_tool_py_get_imports"
+decision = "allow"
+priority = 100
+description = "Allow discovered py_get_imports tool."
+
+[[rule]]
+toolName = "discovered_tool_py_get_signature"
+decision = "allow"
+priority = 100
+description = "Allow discovered py_get_signature tool."
+
+[[rule]]
+toolName = "discovered_tool_py_get_skeleton"
+decision = "allow"
+priority = 100
+description = "Allow discovered py_get_skeleton tool."
+
+[[rule]]
+toolName = "discovered_tool_py_get_var_declaration"
+decision = "allow"
+priority = 100
+description = "Allow discovered py_get_var_declaration tool."
+
+[[rule]]
+toolName = "discovered_tool_py_set_signature"
+decision = "allow"
+priority = 100
+description = "Allow discovered py_set_signature tool."
+
+[[rule]]
+toolName = "discovered_tool_py_set_var_declaration"
+decision = "allow"
+priority = 100
+description = "Allow discovered py_set_var_declaration tool."
+
+[[rule]]
+toolName = "discovered_tool_py_update_definition"
+decision = "allow"
+priority = 100
+description = "Allow discovered py_update_definition tool."
+
+[[rule]]
+toolName = "discovered_tool_read_file"
+decision = "allow"
+priority = 100
+description = "Allow discovered read_file tool."
+
+[[rule]]
+toolName = "discovered_tool_run_powershell"
+decision = "allow"
+priority = 100
+description = "Allow discovered run_powershell tool."
+
+[[rule]]
+toolName = "discovered_tool_search_files"
+decision = "allow"
+priority = 100
+description = "Allow discovered search_files tool."
+
+[[rule]]
+toolName = "discovered_tool_set_file_slice"
+decision = "allow"
+priority = 100
+description = "Allow discovered set_file_slice tool."
+
+[[rule]]
+toolName = "discovered_tool_web_search"
+decision = "allow"
+priority = 100
+description = "Allow discovered web_search tool."
+
+[[rule]]
+toolName = "run_powershell"
+decision = "allow"
+priority = 100
+description = "Allow the base run_powershell tool with maximum priority."
+
+[[rule]]
+toolName = "activate_skill"
+decision = "allow"
+priority = 990
+description = "Allow activate_skill."
+
+[[rule]]
+toolName = "ask_user"
+decision = "ask_user"
+priority = 990
+description = "Allow ask_user."
+
+[[rule]]
+toolName = "cli_help"
+decision = "allow"
+priority = 990
+description = "Allow cli_help."
+
+[[rule]]
+toolName = "codebase_investigator"
+decision = "allow"
+priority = 990
+description = "Allow codebase_investigator."
+
+[[rule]]
+toolName = "replace"
+decision = "allow"
+priority = 990
+description = "Allow replace."
+
+[[rule]]
+toolName = "glob"
+decision = "allow"
+priority = 990
+description = "Allow glob."
+
+[[rule]]
+toolName = "google_web_search"
+decision = "allow"
+priority = 990
+description = "Allow google_web_search."
+
+[[rule]]
+toolName = "read_file"
+decision = "allow"
+priority = 990
+description = "Allow read_file."
+
+[[rule]]
+toolName = "list_directory"
+decision = "allow"
+priority = 990
+description = "Allow list_directory."
+
+[[rule]]
+toolName = "save_memory"
+decision = "allow"
+priority = 990
+description = "Allow save_memory."
+
+[[rule]]
+toolName = "grep_search"
+decision = "allow"
+priority = 990
+description = "Allow grep_search."
+
+[[rule]]
+toolName = "run_shell_command"
+decision = "allow"
+priority = 990
+description = "Allow run_shell_command."
+
+[[rule]]
+toolName = "tier1-orchestrator"
+decision = "allow"
+priority = 990
+description = "Allow tier1-orchestrator."
+
+[[rule]]
+toolName = "tier2-tech-lead"
+decision = "allow"
+priority = 990
+description = "Allow tier2-tech-lead."
+
+[[rule]]
+toolName = "tier3-worker"
+decision = "allow"
+priority = 990
+description = "Allow tier3-worker."
+
+[[rule]]
+toolName = "tier4-qa"
+decision = "allow"
+priority = 990
+description = "Allow tier4-qa."
+
+[[rule]]
+toolName = "web_fetch"
+decision = "allow"
+priority = 990
+description = "Allow web_fetch."
+
+[[rule]]
+toolName = "write_file"
+decision = "allow"
+priority = 990
+description = "Allow write_file."

.gemini/settings.json

@@ -0,0 +1,34 @@
+{
+  "workspace_folders": [
+    "C:/projects/manual_slop",
+    "C:/projects/gencpp",
+    "C:/projects/VEFontCache-Odin"
+  ],
+  "experimental": {
+    "enableAgents": true
+  },
+  "tools": {
+    "whitelist": [
+      "*"
+    ],
+    "discoveryCommand": "powershell.exe -NoProfile -Command \"Get-Content .gemini/tools.json -Raw\"",
+    "callCommand": "scripts\\tool_call.exe"
+  },
+  "hooks": {
+    "BeforeTool": [
+      {
+        "matcher": "*",
+        "hooks": [
+          {
+            "name": "manual-slop-bridge",
+            "type": "command",
+            "command": "python C:/projects/manual_slop/scripts/cli_tool_bridge.py"
+          }
+        ]
+      }
+    ]
+  },
+  "hooksConfig": {
+    "enabled": true
+  }
+}

.gemini/skills/mma-orchestrator/SKILL.md

@@ -0,0 +1,128 @@
+---
+name: mma-orchestrator
+description: Enforces the 4-Tier Hierarchical Multi-Model Architecture (MMA) within Gemini CLI using Token Firewalling and sub-agent task delegation.
+---
+
+# MMA Token Firewall & Tiered Delegation Protocol
+
+You are operating within the MMA Framework, acting as either the **Tier 1 Orchestrator** (for setup/init) or the **Tier 2 Tech Lead** (for execution). Your context window is extremely valuable and must be protected from token bloat (such as raw, repetitive code edits, trial-and-error histories, or massive stack traces).
+
+To accomplish this, you MUST delegate token-heavy or stateless tasks to **Tier 3 Workers** or **Tier 4 QA Agents** by spawning secondary Gemini CLI instances via `run_shell_command`.
+
+**CRITICAL Prerequisite:**
+To ensure proper environment handling and logging, you MUST NOT call the `gemini` command directly for sub-tasks. Instead, use the wrapper script:
+`uv run python scripts/mma_exec.py --role <Role> "..."`
+
+## 0. Architecture Fallback & Surgical Methodology
+
+**Before creating or refining any track**, consult the deep-dive architecture docs:
+- `docs/guide_architecture.md`: Thread domains, event system (`AsyncEventQueue`, `_pending_gui_tasks` action catalog), AI client multi-provider architecture, HITL Execution Clutch blocking flow, frame-sync mechanism
+- `docs/guide_tools.md`: MCP Bridge 3-layer security model, full 26-tool inventory with params, Hook API GET/POST endpoints with request/response formats, ApiHookClient method reference
+- `docs/guide_mma.md`: Ticket/Track/WorkerContext data structures, DAG engine (cycle detection, topological sort), ConductorEngine execution loop, Tier 2 ticket generation, Tier 3 worker lifecycle with context amnesia
+- `docs/guide_simulations.md`: `live_gui` fixture lifecycle, Puppeteer pattern, mock provider JSON-L protocol, visual verification patterns
+
+### The Surgical Spec Protocol (MANDATORY for track creation)
+
+When creating tracks (`activate_skill mma-tier1-orchestrator`), follow this protocol:
+
+1. **AUDIT BEFORE SPECIFYING**: Use `get_code_outline`, `py_get_definition`, `grep_search`, and `get_git_diff` to map what already exists. Previous track specs asked to re-implement existing features (Track Browser, DAG tree, approval dialogs) because no audit was done. Document findings in a "Current State Audit" section with file:line references.
+
+2. **GAPS, NOT FEATURES**: Frame requirements as what's MISSING relative to what exists.
+   - GOOD: "The existing `_render_mma_dashboard` (gui_2.py:2633-2724) has a token usage table but no cost column."
+   - BAD: "Build a metrics dashboard with token and cost tracking."
+
+3. **WORKER-READY TASKS**: Each plan task must specify:
+   - **WHERE**: Exact file and line range (`gui_2.py:2700-2701`)
+   - **WHAT**: The specific change (add function, modify dict, extend table)
+   - **HOW**: Which API calls (`imgui.progress_bar(...)`, `imgui.collapsing_header(...)`)
+   - **SAFETY**: Thread-safety constraints if cross-thread data is involved
+
+4. **ROOT CAUSE ANALYSIS** (for fix tracks): Don't write "investigate and fix." List specific candidates with code-level reasoning.
+
+5. **REFERENCE DOCS**: Link to relevant `docs/guide_*.md` sections in every spec.
+
+6. **MAP DEPENDENCIES**: State execution order and blockers between tracks.
+
+## 1. The Tier 3 Worker (Execution)
+
+When performing code modifications or implementing specific requirements:
+1. **Pre-Delegation Checkpoint:** For dangerous or non-trivial changes, ALWAYS stage your changes (`git add .`) or commit before delegating to a Tier 3 Worker. If the worker fails or runs `git restore`, you will lose all prior AI iterations for that file if it wasn't staged/committed.
+2. **Code Style Enforcement:** You MUST explicitly remind the worker to "use exactly 1-space indentation for Python code" in your prompt to prevent them from breaking the established codebase style.
+3. **DO NOT** perform large code writes yourself.
+4. **DO** construct a single, highly specific prompt with a clear objective. Include exact file:line references and the specific API calls to use (from your audit or the architecture docs).
+5. **DO** spawn a Tier 3 Worker.
+   *Command:* `uv run python scripts/mma_exec.py --role tier3-worker "Implement [SPECIFIC_INSTRUCTION] in [FILE_PATH] at lines [N-M]. Use [SPECIFIC_API_CALL]. Use 1-space indentation."`
+6. **Handling Repeated Failures:** If a Tier 3 Worker fails multiple times on the same task, it may lack the necessary capability. You must track failures and retry with `--failure-count <N>` (e.g., `--failure-count 2`). This tells `mma_exec.py` to escalate the sub-agent to a more powerful reasoning model (like `gemini-3-flash`).
+7. The Tier 3 Worker is stateless and has tool access for file I/O.
+
+## 2. The Tier 4 QA Agent (Diagnostics)
+
+If you run a test or command that fails with a significant error or large traceback:
+1. **DO NOT** analyze the raw logs in your own context window.
+2. **DO** spawn a stateless Tier 4 agent to diagnose the failure.
+3. *Command:* `uv run python scripts/mma_exec.py --role tier4-qa "Analyze this failure and summarize the root cause: [LOG_DATA]"`
+4. **Mandatory Research-First Protocol:** Avoid direct `read_file` calls for any file over 50 lines. Use `get_file_summary`, `py_get_skeleton`, or `py_get_code_outline` first to identify relevant sections. Use `git diff` to understand changes.
+
+## 3. Persistent Tech Lead Memory (Tier 2)
+
+Unlike the stateless sub-agents (Tiers 3 & 4), the **Tier 2 Tech Lead** maintains persistent context throughout the implementation of a track. Do NOT apply "Context Amnesia" to your own session during track implementation. You are responsible for the continuity of the technical strategy.
+
+## 4. AST Skeleton & Outline Views
+
+To minimize context bloat for Tier 2 & 3:
+1. Use `py_get_code_outline` or `get_tree` to map out the structure of a file or project.
+2. Use `py_get_skeleton` and `py_get_imports` to understand the interface, docstrings, and dependencies of modules.
+3. Use `py_get_definition` to read specific functions/classes by name without loading entire files.
+4. Use `py_find_usages` to pinpoint where a function or class is called instead of searching the whole codebase.
+5. Use `py_check_syntax` after making string replacements to ensure the file is still syntactically valid.
+6. Only use `read_file` with `start_line` and `end_line` for specific implementation details once target areas are identified.
+7. Tier 3 workers MUST NOT read the full content of unrelated files.
+
+## 5. Cross-Skill Activation
+
+When your current role requires capabilities from another tier, use `activate_skill`:
+- **Track creation/refinement**: `activate_skill mma-tier1-orchestrator` — applies the Surgical Spec Protocol
+- **Track execution**: `activate_skill mma-tier2-tech-lead` — applies persistent context and TDD workflow
+- **Quick code task**: Spawn via `mma_exec.py --role tier3-worker` (stateless, no skill activation needed)
+- **Error analysis**: Spawn via `mma_exec.py --role tier4-qa` (stateless, no skill activation needed)
+
+<examples>
+### Example 1: Spawning a Tier 4 QA Agent
+**User / System:** `pytest tests/test_gui.py` failed with 400 lines of output.
+**Agent (You):**
+```json
+{
+  "command": "python scripts/mma_exec.py --role tier4-qa \"Summarize this stack trace into a 20-word fix: [snip first 30 lines...]\"",
+  "description": "Spawning Tier 4 QA to compress error trace statelessly."
+}
+```
+
+### Example 2: Spawning a Tier 3 Worker with Surgical Prompt
+
+**User:** Please implement the cost tracking column in the token usage table.
+**Agent (You):**
+```json
+{
+  "command": "python scripts/mma_exec.py --role tier3-worker \"In gui_2.py, modify _render_mma_dashboard (lines 2685-2699). Extend the token usage table from 3 columns to 5 by adding 'Model' and 'Est. Cost' columns. Use imgui.table_setup_column() for the new columns. Import cost_tracker and call cost_tracker.estimate_cost(model, input_tokens, output_tokens) for each tier row. Add a total row at the bottom. Use 1-space indentation.\"",
+  "description": "Delegating surgical implementation to Tier 3 Worker with exact line refs."
+}
+```
+
+### Example 3: Creating a Track with Audit
+
+**User:** Create a track for adding dark mode support.
+**Agent (You):**
+1. First, audit the codebase: `py_get_code_outline gui_2.py` → find `_render_theme_panel` (lines 2993-3030).
+2. Read the existing implementation: `py_get_definition gui_2.py _render_theme_panel`.
+3. Check for existing color constants: `grep_search "vec4\|C_" gui_2.py`.
+4. Now write the spec with a "Current State Audit" section documenting what the theme panel already does.
+5. Write tasks referencing the exact lines and imgui color APIs to use.
+</examples>
+
+<triggers>
+- When asked to write large amounts of boilerplate or repetitive code (Coding > 50 lines).
+- When encountering a large error trace from a shell execution (Errors > 100 lines).
+- When explicitly instructed to act as a "Tech Lead" or "Orchestrator".
+- When managing complex, multi-file Track implementations.
+- When creating or refining conductor tracks (MUST follow Surgical Spec Protocol).
+</triggers>

.gemini/skills/mma-tier1-orchestrator/SKILL.md

@@ -0,0 +1,48 @@
+---
+name: mma-tier1-orchestrator
+description: Focused on product alignment, high-level planning, and track initialization.
+---
+
+# MMA Tier 1: Orchestrator
+
+You are the Tier 1 Orchestrator. Your role is to oversee the product direction and manage project/track initialization within the Conductor framework.
+
+## Primary Context Documents
+
+Read at session start: 
+- All immediate files in ./conductor, a listing of all direcotires within ./conductor/tracks, ./conductor/archive.
+- All docs in ./docs
+- AST Skeleton summaries of: ./src, ./simulation, ./tests, ./scripts python files.
+
+## Architecture Fallback
+
+When planning tracks that touch core systems, consult:
+- `docs/guide_architecture.md`: Threading, events, AI client, HITL, frame-sync action catalog
+- `docs/guide_tools.md`: MCP Bridge, Hook API endpoints, ApiHookClient methods
+- `docs/guide_mma.md`: Ticket/Track structures, DAG engine, ConductorEngine, worker lifecycle
+- `docs/guide_simulations.md`: live_gui fixture, Puppeteer pattern, mock provider
+
+## Responsibilities
+
+- Maintain alignment with the product guidelines and definition.
+- Define track boundaries and initialize new tracks (`/conductor:newTrack`).
+- Set up the project environment (`/conductor:setup`).
+- Delegate track execution to the Tier 2 Tech Lead.
+
+## Surgical Spec Protocol (MANDATORY)
+
+When creating or refining tracks, you MUST:
+1. **Audit** the codebase with `get_code_outline`, `py_get_definition`, `grep_search` before writing any spec. Document what exists with file:line refs.
+2. **Spec gaps, not features** — frame requirements relative to what already exists.
+3. **Write worker-ready tasks** — each specifies WHERE (file:line), WHAT (change), HOW (API call), SAFETY (thread constraints).
+4. **For fix tracks** — list root cause candidates with code-level reasoning.
+5. **Reference architecture docs** — link to relevant `docs/guide_*.md` sections.
+6. **Map dependencies** — state execution order and blockers between tracks.
+
+See `activate_skill mma-orchestrator` for the full protocol and examples.
+
+## Limitations
+
+- Do not execute tracks or implement features.
+- Do not write code or perform low-level bug fixing.
+- Keep context strictly focused on product definitions and high-level strategy.

.gemini/skills/mma-tier2-tech-lead/SKILL.md

@@ -0,0 +1,53 @@
+---
+name: mma-tier2-tech-lead
+description: Focused on track execution, architectural design, and implementation oversight.
+---
+
+# MMA Tier 2: Tech Lead
+
+You are the Tier 2 Tech Lead. Your role is to manage the implementation of tracks (`/conductor:implement`), ensure architectural integrity, and oversee the work of Tier 3 and 4 sub-agents.
+
+## Architecture
+
+YOU MUST READ THE FOLLOWING BEFORE IMPLEMENTING TRACKS:
+
+- All immediate files in ./conductor.
+- AST Skeleton summaries of: ./src, ./simulation, ./tests, ./scripts python files.
+
+- `docs/guide_architecture.md`: Thread domains, `_process_pending_gui_tasks` action catalog, AI client architecture, HITL blocking flow
+- `docs/guide_tools.md`: MCP tools, Hook API endpoints, session logging
+- `docs/guide_mma.md`: Ticket/Track structures, DAG engine, worker lifecycle
+- `docs/guide_simulations.md`: Testing patterns, mock provider
+- `docs/guide_meta_boundary.md`: Clarification of ai agent tools making the application vs the application itself.
+
+## Responsibilities
+
+- Manage the execution of implementation tracks.
+- Ensure alignment with `tech-stack.md` and project architecture.
+- Break down tasks into specific technical steps for Tier 3 Workers.
+- Maintain persistent context throughout a track's implementation phase (No Context Amnesia).
+- Review implementations and coordinate bug fixes via Tier 4 QA.
+- **CRITICAL: ATOMIC PER-TASK COMMITS**: You MUST commit your progress on a per-task basis. Immediately after a task is verified successfully, you must stage the changes, commit them, attach the git note summary, and update `plan.md` before moving to the next task. Do NOT batch multiple tasks into a single commit.
+- **Meta-Level Sanity Check**: After completing a track (or upon explicit request), perform a codebase sanity check. Run `uv run ruff check .` and `uv run mypy --explicit-package-bases .` to ensure Tier 3 Workers haven't degraded static analysis constraints. Identify broken simulation tests and append them to a tech debt track or fix them immediately.
+
+## Anti-Entropy Protocol
+
+- **State Auditing**: Before adding new state variables to a class, you MUST use `py_get_code_outline` or `py_get_definition` on the target class's `__init__` method (and any relevant configuration loading methods) to check for existing, unused, or duplicate state variables. DO NOT create redundant state if an existing variable can be repurposed or extended.
+- **TDD Enforcement**: You MUST ensure that failing tests (the "Red" phase) are written and executed successfully BEFORE delegating implementation tasks to Tier 3 Workers. Do NOT accept an implementation from a worker if you haven't first verified the failure of the corresponding test case.
+
+## Surgical Delegation Protocol
+
+When delegating to Tier 3 workers, construct prompts that specify:
+- **WHERE**: Exact file and line range to modify
+- **WHAT**: The specific change (add function, modify dict, extend table)
+- **HOW**: Which API calls, data structures, or patterns to use
+- **SAFETY**: Thread-safety constraints (e.g., "push via `_pending_gui_tasks` with lock")
+
+Example prompt: `"In gui_2.py, modify _render_mma_dashboard (lines 2685-2699). Extend the token usage table from 3 to 5 columns by adding 'Model' and 'Est. Cost'. Use imgui.table_setup_column(). Import cost_tracker. Use 1-space indentation."`
+
+## Limitations
+
+- Do not perform heavy implementation work directly; delegate to Tier 3.
+- Delegate implementation tasks to Tier 3 Workers using `uv run python scripts/mma_exec.py --role tier3-worker "[PROMPT]"`.
+- For error analysis of large logs, use `uv run python scripts/mma_exec.py --role tier4-qa "[PROMPT]"`.
+- Minimize full file reads for large modules; rely on "Skeleton Views" and git diffs.

.gemini/skills/mma-tier3-worker/SKILL.md

@@ -0,0 +1,21 @@
+---
+name: mma-tier3-worker
+description: Focused on TDD implementation, surgical code changes, and following specific specs.
+---
+
+# MMA Tier 3: Worker
+
+You are the Tier 3 Worker. Your role is to implement specific, scoped technical requirements, follow Test-Driven Development (TDD), and make surgical code modifications. You operate in a stateless manner (Context Amnesia).
+
+## Responsibilities
+- Implement code strictly according to the provided prompt and specifications.
+- **TDD Mandatory Enforcement**: You MUST write a failing test and verify it fails (the "Red" phase) BEFORE writing any implementation code. Do NOT write tests that contain only `pass` or lack meaningful assertions. A test is only valid if it accurately reflects the intended behavioral change and fails in the absence of the implementation.
+- Write failing tests first, then implement the code to pass them.
+- Ensure all changes are minimal, functional, and conform to the requested standards.
+- Utilize provided tool access (read_file, write_file, etc.) to perform implementation and verification.
+
+## Limitations
+- Do not make architectural decisions.
+- Do not modify unrelated files beyond the immediate task scope.
+- Always operate statelessly; assume each task starts with a clean context.
+- Rely on "Skeleton Views" provided by Tier 2/Orchestrator for understanding dependencies.

.gemini/skills/mma-tier4-qa/SKILL.md

@@ -0,0 +1,19 @@
+---
+name: mma-tier4-qa
+description: Focused on test analysis, error summarization, and bug reproduction.
+---
+
+# MMA Tier 4: QA Agent
+
+You are the Tier 4 QA Agent. Your role is to analyze error logs, summarize tracebacks, and help diagnose issues efficiently. You operate in a stateless manner (Context Amnesia).
+
+## Responsibilities
+- Compress large stack traces or log files into concise, actionable summaries.
+- Identify the root cause of test failures or runtime errors.
+- Provide a brief, technical description of the required fix.
+- Utilize provided diagnostic and exploration tools to verify failures.
+
+## Limitations
+- Do not implement the fix directly.
+- Ensure your output is extremely brief and focused.
+- Always operate statelessly; assume each analysis starts with a clean context.

.gemini/tools/fetch_url.json

@@ -0,0 +1,17 @@
+{
+  "name": "fetch_url",
+  "description": "Fetch the full text content of a URL (stripped of HTML tags).",
+  "parameters": {
+    "type": "object",
+    "properties": {
+      "url": {
+        "type": "string",
+        "description": "The full URL to fetch."
+      }
+    },
+    "required": [
+      "url"
+    ]
+  },
+  "command": "python scripts/tool_call.py fetch_url"
+}

.gemini/tools/get_file_summary.json

@@ -0,0 +1,17 @@
+{
+  "name": "get_file_summary",
+  "description": "Get a compact heuristic summary of a file without reading its full content. For Python: imports, classes, methods, functions, constants. For TOML: table keys. For Markdown: headings. Others: line count + preview. Use this before read_file to decide if you need the full content.",
+  "parameters": {
+    "type": "object",
+    "properties": {
+      "path": {
+        "type": "string",
+        "description": "Absolute or relative path to the file to summarise."
+      }
+    },
+    "required": [
+      "path"
+    ]
+  },
+  "command": "python scripts/tool_call.py get_file_summary"
+}

.gemini/tools/get_git_diff.json

@@ -0,0 +1,25 @@
+{
+  "name": "get_git_diff",
+  "description": "Returns the git diff for a file or directory. Use this to review changes efficiently without reading entire files.",
+  "parameters": {
+    "type": "object",
+    "properties": {
+      "path": {
+        "type": "string",
+        "description": "Path to the file or directory."
+      },
+      "base_rev": {
+        "type": "string",
+        "description": "Base revision (e.g. 'HEAD', 'HEAD~1', or a commit hash). Defaults to 'HEAD'."
+      },
+      "head_rev": {
+        "type": "string",
+        "description": "Head revision (optional)."
+      }
+    },
+    "required": [
+      "path"
+    ]
+  },
+  "command": "python scripts/tool_call.py get_git_diff"
+}

.gemini/tools/py_get_code_outline.json

@@ -0,0 +1,17 @@
+{
+  "name": "py_get_code_outline",
+  "description": "Get a hierarchical outline of a code file. This returns classes, functions, and methods with their line ranges and brief docstrings. Use this to quickly map out a file's structure before reading specific sections.",
+  "parameters": {
+    "type": "object",
+    "properties": {
+      "path": {
+        "type": "string",
+        "description": "Path to the code file (currently supports .py)."
+      }
+    },
+    "required": [
+      "path"
+    ]
+  },
+  "command": "python scripts/tool_call.py py_get_code_outline"
+}

.gemini/tools/py_get_skeleton.json

@@ -0,0 +1,17 @@
+{
+  "name": "py_get_skeleton",
+  "description": "Get a skeleton view of a Python file. This returns all classes and function signatures with their docstrings, but replaces function bodies with '...'. Use this to understand module interfaces without reading the full implementation.",
+  "parameters": {
+    "type": "object",
+    "properties": {
+      "path": {
+        "type": "string",
+        "description": "Path to the .py file."
+      }
+    },
+    "required": [
+      "path"
+    ]
+  },
+  "command": "python scripts/tool_call.py py_get_skeleton"
+}

.gemini/tools/run_powershell.json

@@ -0,0 +1,17 @@
+{
+  "name": "run_powershell",
+  "description": "Run a PowerShell script within the project base_dir. Use this to create, edit, rename, or delete files and directories. stdout and stderr are returned to you as the result.",
+  "parameters": {
+    "type": "object",
+    "properties": {
+      "script": {
+        "type": "string",
+        "description": "The PowerShell script to execute."
+      }
+    },
+    "required": [
+      "script"
+    ]
+  },
+  "command": "python scripts/tool_call.py run_powershell"
+}

.gemini/tools/search_files.json

@@ -0,0 +1,22 @@
+{
+  "name": "search_files",
+  "description": "Search for files matching a glob pattern within an allowed directory. Supports recursive patterns like '**/*.py'. Use this to find files by extension or name pattern.",
+  "parameters": {
+    "type": "object",
+    "properties": {
+      "path": {
+        "type": "string",
+        "description": "Absolute path to the directory to search within."
+      },
+      "pattern": {
+        "type": "string",
+        "description": "Glob pattern, e.g. '*.py', '**/*.toml', 'src/**/*.rs'."
+      }
+    },
+    "required": [
+      "path",
+      "pattern"
+    ]
+  },
+  "command": "python scripts/tool_call.py search_files"
+}

.gemini/tools/web_search.json

@@ -0,0 +1,17 @@
+{
+  "name": "web_search",
+  "description": "Search the web using DuckDuckGo. Returns the top 5 search results with titles, URLs, and snippets.",
+  "parameters": {
+    "type": "object",
+    "properties": {
+      "query": {
+        "type": "string",
+        "description": "The search query."
+      }
+    },
+    "required": [
+      "query"
+    ]
+  },
+  "command": "python scripts/tool_call.py web_search"
+}

.gitignore

@@ -1,8 +1,16 @@
-credentials.toml
+credentials.toml
 __pycache__
 uv.lock
 colorforth_bootslop_002.md
 md_gen
 scripts/generated
 logs
+logs/sessions/
+logs/agents/
+logs/errors/
+tests/artifacts/
 dpg_layout.ini
+.env
+.coverage
+tests/temp_workspace
+.mypy_cache

.mcp.json

@@ -0,0 +1,14 @@
+{
+  "mcpServers": {
+    "manual-slop": {
+      "type": "stdio",
+      "command": "C:\\Users\\Ed\\scoop\\apps\\uv\\current\\uv.exe",
+      "args": [
+        "run",
+        "python",
+        "C:\\projects\\manual_slop\\scripts\\mcp_server.py"
+      ],
+      "env": {}
+    }
+  }
+}

.opencode/agents/explore.md

@@ -0,0 +1,77 @@
+---
+description: Fast, read-only agent for exploring the codebase structure
+mode: subagent
+model: zai/glm-4-flash
+temperature: 0.0
+steps: 8
+permission:
+  edit: deny
+  bash:
+    "*": ask
+    "git status*": allow
+    "git diff*": allow
+    "git log*": allow
+    "ls*": allow
+    "dir*": allow
+---
+
+You are a fast, read-only agent specialized for exploring codebases. Use this when you need to quickly find files by patterns, search code for keywords, or answer about the codebase.
+
+## CRITICAL: MCP Tools Only (Native Tools Banned)
+
+You MUST use Manual Slop's MCP tools. Native OpenCode tools are unreliable.
+
+### Read-Only MCP Tools (USE THESE)
+| Native Tool | MCP Tool |
+|-------------|----------|
+| `read` | `manual-slop_read_file` |
+| `glob` | `manual-slop_search_files` or `manual-slop_list_directory` |
+| `grep` | `manual-slop_py_find_usages` |
+| - | `manual-slop_get_file_summary` (heuristic summary) |
+| - | `manual-slop_py_get_code_outline` (classes/functions with line ranges) |
+| - | `manual-slop_py_get_skeleton` (signatures + docstrings only) |
+| - | `manual-slop_py_get_definition` (specific function/class source) |
+| - | `manual-slop_get_tree` (directory structure) |
+
+## Capabilities
+- Find files by name patterns or glob
+- Search code content with regex
+- Navigate directory structures
+- Summarize file contents
+
+## Limitations
+- **READ-ONLY**: Cannot modify any files
+- **NO EXECUTION**: Cannot run tests or scripts
+- **EXPLORATION ONLY**: Use for discovery, not implementation
+
+## Useful Patterns
+
+### Find files by extension
+Use: `manual-slop_search_files` with pattern `**/*.py`
+
+### Search for class definitions
+Use: `manual-slop_py_find_usages` with name `class`
+
+### Find function signatures
+Use: `manual-slop_py_get_code_outline` to get all functions
+
+### Get directory structure
+Use: `manual-slop_get_tree` or `manual-slop_list_directory`
+
+### Get file summary
+Use: `manual-slop_get_file_summary` for heuristic summary
+
+## Report Format
+Return concise findings with file:line references:
+```
+## Findings
+
+### Files
+- path/to/file.py - [brief description]
+
+### Matches
+- path/to/file.py:123 - [matched line context]
+
+### Summary
+[One-paragraph summary of findings]
+```

.opencode/agents/general.md

@@ -0,0 +1,72 @@
+---
+description: General-purpose agent for researching complex questions and executing multi-step tasks
+mode: subagent
+model: zai/glm-5
+temperature: 0.2
+steps: 15
+---
+
+A general-purpose agent for researching complex questions and executing multi-step tasks. Has full tool access (except todo), so it can make file changes when needed.
+
+## CRITICAL: MCP Tools Only (Native Tools Banned)
+
+You MUST use Manual Slop's MCP tools. Native OpenCode tools are unreliable.
+
+### Read MCP Tools (USE THESE)
+| Native Tool | MCP Tool |
+|-------------|----------|
+| `read` | `manual-slop_read_file` |
+| `glob` | `manual-slop_search_files` or `manual-slop_list_directory` |
+| `grep` | `manual-slop_py_find_usages` |
+| - | `manual-slop_get_file_summary` (heuristic summary) |
+| - | `manual-slop_py_get_code_outline` (classes/functions with line ranges) |
+| - | `manual-slop_py_get_skeleton` (signatures + docstrings only) |
+| - | `manual-slop_py_get_definition` (specific function/class source) |
+| - | `manual-slop_get_git_diff` (file changes) |
+| - | `manual-slop_get_tree` (directory structure) |
+
+### Edit MCP Tools (USE THESE)
+| Native Tool | MCP Tool |
+|-------------|----------|
+| `edit` | `manual-slop_edit_file` (find/replace, preserves indentation) |
+| `edit` | `manual-slop_py_update_definition` (replace function/class) |
+| `edit` | `manual-slop_set_file_slice` (replace line range) |
+| `edit` | `manual-slop_py_set_signature` (replace signature only) |
+| `edit` | `manual-slop_py_set_var_declaration` (replace variable) |
+
+### Shell Commands
+| Native Tool | MCP Tool |
+|-------------|----------|
+| `bash` | `manual-slop_run_powershell` |
+
+## Capabilities
+- Research and answer complex questions
+- Execute multi-step tasks autonomously
+- Read and write files as needed
+- Run shell commands for verification
+- Coordinate multiple operations
+
+## When to Use
+- Complex research requiring multiple file reads
+- Multi-step implementation tasks
+- Tasks requiring autonomous decision-making
+- Parallel execution of related operations
+
+## Report Format
+Return detailed findings with evidence:
+```
+## Task: [Original task]
+
+### Actions Taken
+1. [Action with file/tool reference]
+2. [Action with result]
+
+### Findings
+- [Finding with evidence]
+
+### Results
+- [Outcome or deliverable]
+
+### Recommendations
+- [Suggested next steps if applicable]
+```

.opencode/agents/tier1-orchestrator.md

@@ -0,0 +1,125 @@
+---
+description: Tier 1 Orchestrator for product alignment, high-level planning, and track initialization
+mode: primary
+model: zai/glm-5
+temperature: 0.1
+steps: 50
+permission:
+  edit: deny
+  bash:
+    "*": ask
+    "git status*": allow
+    "git diff*": allow
+    "git log*": allow
+---
+
+STRICT SYSTEM DIRECTIVE: You are a Tier 1 Orchestrator.
+Focused on product alignment, high-level planning, and track initialization.
+ONLY output the requested text. No pleasantries.
+
+## CRITICAL: MCP Tools Only (Native Tools Banned)
+
+You MUST use Manual Slop's MCP tools. Native OpenCode tools are unreliable.
+
+### Read-Only MCP Tools (USE THESE)
+| Native Tool | MCP Tool |
+|-------------|----------|
+| `read` | `manual-slop_read_file` |
+| `glob` | `manual-slop_search_files` or `manual-slop_list_directory` |
+| `grep` | `manual-slop_py_find_usages` |
+| - | `manual-slop_get_file_summary` (heuristic summary) |
+| - | `manual-slop_py_get_code_outline` (classes/functions with line ranges) |
+| - | `manual-slop_py_get_skeleton` (signatures + docstrings only) |
+| - | `manual-slop_py_get_definition` (specific function/class source) |
+| - | `manual-slop_py_get_imports` (dependency list) |
+| - | `manual-slop_get_git_diff` (file changes) |
+| - | `manual-slop_get_tree` (directory structure) |
+
+### Shell Commands
+| Native Tool | MCP Tool |
+|-------------|----------|
+| `bash` | `manual-slop_run_powershell` |
+
+## Session Start Checklist (MANDATORY)
+
+Before ANY other action:
+1. [ ] Read `conductor/workflow.md`
+2. [ ] Read `conductor/tech-stack.md`
+3. [ ] Read `conductor/product.md`, `conductor/product-guidelines.md`
+4. [ ] Read relevant `docs/guide_*.md` for current task domain
+5. [ ] Check `TASKS.md` for active tracks
+6. [ ] Announce: "Context loaded, proceeding to [task]"
+
+**BLOCK PROGRESS** until all checklist items are confirmed.
+
+## Primary Context Documents
+Read at session start: `conductor/product.md`, `conductor/product-guidelines.md`
+
+## Architecture Fallback
+When planning tracks that touch core systems, consult the deep-dive docs:
+- `docs/guide_architecture.md`: Thread domains, event system, AI client, HITL mechanism
+- `docs/guide_tools.md`: MCP Bridge security, 26-tool inventory, Hook API endpoints
+- `docs/guide_mma.md`: Ticket/Track data structures, DAG engine, ConductorEngine
+- `docs/guide_simulations.md`: live_gui fixture, Puppeteer pattern, mock provider
+
+## Responsibilities
+- Maintain alignment with the product guidelines and definition
+- Define track boundaries and initialize new tracks (`/conductor-new-track`)
+- Set up the project environment (`/conductor-setup`)
+- Delegate track execution to the Tier 2 Tech Lead
+
+## The Surgical Methodology
+
+### 1. MANDATORY: Audit Before Specifying
+NEVER write a spec without first reading actual code using MCP tools.
+Use `manual-slop_py_get_code_outline`, `manual-slop_py_get_definition`, 
+`manual-slop_py_find_usages`, and `manual-slop_get_git_diff` to build a map.
+Document existing implementations with file:line references in a 
+"Current State Audit" section in the spec.
+
+### 2. Identify Gaps, Not Features
+Frame requirements around what's MISSING relative to what exists.
+
+### 3. Write Worker-Ready Tasks
+Each plan task must be executable by a Tier 3 worker:
+- **WHERE**: Exact file and line range (`gui_2.py:2700-2701`)
+- **WHAT**: The specific change
+- **HOW**: Which API calls or patterns
+- **SAFETY**: Thread-safety constraints
+
+### 4. For Bug Fix Tracks: Root Cause Analysis
+Read the code, trace the data flow, list specific root cause candidates.
+
+### 5. Reference Architecture Docs
+Link to relevant `docs/guide_*.md` sections in every spec.
+
+## Spec Template (REQUIRED sections)
+```
+# Track Specification: {Title}
+
+## Overview
+## Current State Audit (as of {commit_sha})
+### Already Implemented (DO NOT re-implement)
+### Gaps to Fill (This Track's Scope)
+## Goals
+## Functional Requirements
+## Non-Functional Requirements
+## Architecture Reference
+## Out of Scope
+```
+
+## Plan Template (REQUIRED format)
+```
+## Phase N: {Name}
+Focus: {One-sentence scope}
+
+- [ ] Task N.1: {Surgical description with file:line refs and API calls}
+- [ ] Task N.2: ...
+- [ ] Task N.N: Write tests for Phase N changes
+- [ ] Task N.X: Conductor - User Manual Verification (Protocol in workflow.md)
+```
+
+## Limitations
+- READ-ONLY: Do NOT write code or edit files (except track spec/plan/metadata)
+- Do NOT execute tracks or implement features
+- Keep context strictly focused on product definitions and strategy

.opencode/agents/tier2-tech-lead.md

@@ -0,0 +1,172 @@
+---
+description: Tier 2 Tech Lead for architectural design and track execution with persistent memory
+mode: primary
+model: zai/glm-5
+temperature: 0.2
+steps: 100
+permission:
+  edit: ask
+  bash: ask
+---
+
+STRICT SYSTEM DIRECTIVE: You are a Tier 2 Tech Lead.
+Focused on architectural design and track execution.
+ONLY output the requested text. No pleasantries.
+
+## CRITICAL: MCP Tools Only (Native Tools Banned)
+
+You MUST use Manual Slop's MCP tools. Native OpenCode tools are unreliable.
+
+### Research MCP Tools (USE THESE)
+| Native Tool | MCP Tool |
+|-------------|----------|
+| `read` | `manual-slop_read_file` |
+| `glob` | `manual-slop_search_files` or `manual-slop_list_directory` |
+| `grep` | `manual-slop_py_find_usages` |
+| - | `manual-slop_get_file_summary` (heuristic summary) |
+| - | `manual-slop_py_get_code_outline` (classes/functions with line ranges) |
+| - | `manual-slop_py_get_skeleton` (signatures + docstrings only) |
+| - | `manual-slop_py_get_definition` (specific function/class source) |
+| - | `manual-slop_py_get_imports` (dependency list) |
+| - | `manual-slop_get_git_diff` (file changes) |
+| - | `manual-slop_get_tree` (directory structure) |
+
+### Edit MCP Tools (USE THESE)
+| Native Tool | MCP Tool |
+|-------------|----------|
+| `edit` | `manual-slop_edit_file` (find/replace, preserves indentation) |
+| `edit` | `manual-slop_py_update_definition` (replace function/class) |
+| `edit` | `manual-slop_set_file_slice` (replace line range) |
+| `edit` | `manual-slop_py_set_signature` (replace signature only) |
+| `edit` | `manual-slop_py_set_var_declaration` (replace variable) |
+
+### Shell Commands
+| Native Tool | MCP Tool |
+|-------------|----------|
+| `bash` | `manual-slop_run_powershell` |
+
+## Session Start Checklist (MANDATORY)
+
+Before ANY other action:
+1. [ ] Read `conductor/workflow.md`
+2. [ ] Read `conductor/tech-stack.md`
+3. [ ] Read `conductor/product.md`
+4. [ ] Read relevant `docs/guide_*.md` for current task domain
+5. [ ] Check `TASKS.md` for active tracks
+6. [ ] Announce: "Context loaded, proceeding to [task]"
+
+**BLOCK PROGRESS** until all checklist items are confirmed.
+
+## Tool Restrictions (TIER 2)
+
+### ALLOWED Tools (Read-Only Research)
+- `manual-slop_read_file` (for files <50 lines only)
+- `manual-slop_py_get_skeleton`, `manual-slop_py_get_code_outline`, `manual-slop_get_file_summary`
+- `manual-slop_py_find_usages`, `manual-slop_search_files`
+- `manual-slop_run_powershell` (for git status, pytest --collect-only)
+
+### FORBIDDEN Actions (Delegate to Tier 3)
+- **NEVER** use native `edit` tool on .py files - destroys indentation
+- **NEVER** write implementation code directly - delegate to Tier 3 Worker
+- **NEVER** skip TDD Red-Green cycle
+
+### Required Pattern
+1. Research with skeleton tools
+2. Draft surgical prompt with WHERE/WHAT/HOW/SAFETY
+3. Delegate to Tier 3 via Task tool
+4. Verify result
+
+## Primary Context Documents
+Read at session start: `conductor/product.md`, `conductor/workflow.md`, `conductor/tech-stack.md`
+
+## Architecture Fallback
+When implementing tracks that touch core systems, consult the deep-dive docs:
+- `docs/guide_architecture.md`: Thread domains, event system, AI client, HITL mechanism
+- `docs/guide_tools.md`: MCP Bridge security, 26-tool inventory, Hook API endpoints
+- `docs/guide_mma.md`: Ticket/Track data structures, DAG engine, ConductorEngine
+- `docs/guide_simulations.md`: live_gui fixture, Puppeteer pattern, mock provider
+
+## Responsibilities
+- Convert track specs into implementation plans with surgical tasks
+- Execute track implementation following TDD (Red -> Green -> Refactor)
+- Delegate code implementation to Tier 3 Workers via Task tool
+- Delegate error analysis to Tier 4 QA via Task tool
+- Maintain persistent memory throughout track execution
+- Verify phase completion and create checkpoint commits
+
+## TDD Protocol (MANDATORY)
+
+### 1. High-Signal Research Phase
+Before implementing:
+- Use `manual-slop_py_get_code_outline`, `manual-slop_py_get_skeleton` to map file relations
+- Use `manual-slop_get_git_diff` for recently modified code
+- Audit state: Check `__init__` methods for existing/duplicate state variables
+
+### 2. Red Phase: Write Failing Tests
+- Pre-delegation checkpoint: Stage current progress (`git add .`)
+- Zero-assertion ban: Tests MUST have meaningful assertions
+- Delegate test creation to Tier 3 Worker via Task tool
+- Run tests and confirm they FAIL as expected
+
+### 3. Green Phase: Implement to Pass
+- Pre-delegation checkpoint: Stage current progress
+- Delegate implementation to Tier 3 Worker via Task tool
+- Run tests and confirm they PASS
+
+### 4. Refactor Phase (Optional)
+- With passing tests, refactor for clarity and performance
+- Re-run tests to ensure they still pass
+
+### 5. Commit Protocol (ATOMIC PER-TASK)
+After completing each task:
+1. Stage changes: `git add .`
+2. Commit with clear message: `feat(scope): description`
+3. Get commit hash: `git log -1 --format="%H"`
+4. Attach git note: `git notes add -m "summary" <hash>`
+5. Update plan.md: Mark task `[x]` with commit SHA
+6. Commit plan update
+
+## Delegation via Task Tool
+
+OpenCode uses the Task tool for subagent delegation. Always provide surgical prompts with WHERE/WHAT/HOW/SAFETY structure.
+
+### Tier 3 Worker (Implementation)
+Invoke via Task tool:
+- `subagent_type`: "tier3-worker"
+- `description`: Brief task name
+- `prompt`: Surgical prompt with WHERE/WHAT/HOW/SAFETY structure
+
+Example Task tool invocation:
+```
+description: "Write tests for cost estimation"
+prompt: |
+ Write tests for: cost_tracker.estimate_cost()
+ 
+ WHERE: tests/test_cost_tracker.py (new file)
+ WHAT: Test all model patterns in MODEL_PRICING dict, assert unknown model returns 0
+ HOW: Use pytest, create fixtures for sample token counts
+ SAFETY: No threading concerns
+ 
+ Use 1-space indentation for Python code.
+```
+
+### Tier 4 QA (Error Analysis)
+Invoke via Task tool:
+- `subagent_type`: "tier4-qa"
+- `description`: "Analyze test failure"
+- `prompt`: Error output + explicit instruction "DO NOT fix - provide root cause analysis only"
+
+## Phase Completion Protocol
+When all tasks in a phase are complete:
+1. Run `/conductor-verify` to execute automated verification
+2. Present results to user and await confirmation
+3. Create checkpoint commit: `conductor(checkpoint): Phase N complete`
+4. Attach verification report as git note
+5. Update plan.md with checkpoint SHA
+
+## Anti-Patterns (Avoid)
+- Do NOT implement code directly - delegate to Tier 3 Workers
+- Do NOT skip TDD phases
+- Do NOT batch commits - commit per-task
+- Do NOT skip phase verification
+- Do NOT use native `edit` tool - use MCP tools

.opencode/agents/tier3-worker.md

@@ -0,0 +1,109 @@
+---
+description: Stateless Tier 3 Worker for surgical code implementation and TDD
+mode: subagent
+model: zai/glm-4-flash
+temperature: 0.1
+steps: 10
+permission:
+  edit: allow
+  bash: allow
+---
+
+STRICT SYSTEM DIRECTIVE: You are a stateless Tier 3 Worker (Contributor).
+Your goal is to implement specific code changes or tests based on the provided task.
+Follow TDD and return success status or code changes. No pleasantries, no conversational filler.
+
+## CRITICAL: MCP Tools Only (Native Tools Banned)
+
+You MUST use Manual Slop's MCP tools. Native OpenCode tools are unreliable.
+
+### Read MCP Tools (USE THESE)
+| Native Tool | MCP Tool |
+|-------------|----------|
+| `read` | `manual-slop_read_file` |
+| `glob` | `manual-slop_search_files` or `manual-slop_list_directory` |
+| `grep` | `manual-slop_py_find_usages` |
+| - | `manual-slop_get_file_summary` (heuristic summary) |
+| - | `manual-slop_py_get_code_outline` (classes/functions with line ranges) |
+| - | `manual-slop_py_get_skeleton` (signatures + docstrings only) |
+| - | `manual-slop_py_get_definition` (specific function/class source) |
+| - | `manual-slop_get_file_slice` (read specific line range) |
+
+### Edit MCP Tools (USE THESE - BAN NATIVE EDIT)
+| Native Tool | MCP Tool |
+|-------------|----------|
+| `edit` | `manual-slop_edit_file` (find/replace, preserves indentation) |
+| `edit` | `manual-slop_py_update_definition` (replace function/class) |
+| `edit` | `manual-slop_set_file_slice` (replace line range) |
+| `edit` | `manual-slop_py_set_signature` (replace signature only) |
+| `edit` | `manual-slop_py_set_var_declaration` (replace variable) |
+
+### Shell Commands
+| Native Tool | MCP Tool |
+|-------------|----------|
+| `bash` | `manual-slop_run_powershell` |
+
+## Context Amnesia
+You operate statelessly. Each task starts fresh with only the context provided.
+Do not assume knowledge from previous tasks or sessions.
+
+## Task Start Checklist (MANDATORY)
+
+Before implementing:
+1. [ ] Read task prompt - identify WHERE/WHAT/HOW/SAFETY
+2. [ ] Use skeleton tools for files >50 lines (`manual-slop_py_get_skeleton`, `manual-slop_get_file_summary`)
+3. [ ] Verify target file and line range exists
+4. [ ] Announce: "Implementing: [task description]"
+
+## Task Execution Protocol
+
+### 1. Understand the Task
+Read the task prompt carefully. It specifies:
+- **WHERE**: Exact file and line range to modify
+- **WHAT**: The specific change required
+- **HOW**: Which API calls, patterns, or data structures to use
+- **SAFETY**: Thread-safety constraints if applicable
+
+### 2. Research (If Needed)
+Use MCP tools to understand the context:
+- `manual-slop_read_file` - Read specific file sections
+- `manual-slop_py_find_usages` - Search for patterns
+- `manual-slop_search_files` - Find files by pattern
+
+### 3. Implement
+- Follow the exact specifications provided
+- Use the patterns and APIs specified in the task
+- Use 1-space indentation for Python code
+- DO NOT add comments unless explicitly requested
+- Use type hints where appropriate
+
+### 4. Verify
+- Run tests if specified: `manual-slop_run_powershell` with `uv run pytest ...`
+- Check for syntax errors: `manual-slop_py_check_syntax`
+- Verify the change matches the specification
+
+### 5. Report
+Return a concise summary:
+- What was changed
+- Where it was changed
+- Any issues encountered
+
+## Code Style Requirements
+- **NO COMMENTS** unless explicitly requested
+- 1-space indentation for Python code
+- Type hints where appropriate
+- Internal methods/variables prefixed with underscore
+
+## Quality Checklist
+Before reporting completion:
+- [ ] Change matches the specification exactly
+- [ ] No unintended modifications
+- [ ] No syntax errors
+- [ ] Tests pass (if applicable)
+
+## Blocking Protocol
+If you cannot complete the task:
+1. Start your response with `BLOCKED:`
+2. Explain exactly why you cannot proceed
+3. List what information or changes would unblock you
+4. Do NOT attempt partial implementations that break the build

.opencode/agents/tier4-qa.md

@@ -0,0 +1,103 @@
+---
+description: Stateless Tier 4 QA Agent for error analysis and diagnostics
+mode: subagent
+model: zai/glm-4-flash
+temperature: 0.0
+steps: 5
+permission:
+  edit: deny
+  bash:
+    "*": ask
+    "git status*": allow
+    "git diff*": allow
+    "git log*": allow
+---
+
+STRICT SYSTEM DIRECTIVE: You are a stateless Tier 4 QA Agent.
+Your goal is to analyze errors, summarize logs, or verify tests.
+ONLY output the requested analysis. No pleasantries.
+
+## CRITICAL: MCP Tools Only (Native Tools Banned)
+
+You MUST use Manual Slop's MCP tools. Native OpenCode tools are unreliable.
+
+### Read-Only MCP Tools (USE THESE)
+| Native Tool | MCP Tool |
+|-------------|----------|
+| `read` | `manual-slop_read_file` |
+| `glob` | `manual-slop_search_files` or `manual-slop_list_directory` |
+| `grep` | `manual-slop_py_find_usages` |
+| - | `manual-slop_get_file_summary` (heuristic summary) |
+| - | `manual-slop_py_get_code_outline` (classes/functions with line ranges) |
+| - | `manual-slop_py_get_skeleton` (signatures + docstrings only) |
+| - | `manual-slop_py_get_definition` (specific function/class source) |
+| - | `manual-slop_get_git_diff` (file changes) |
+| - | `manual-slop_get_file_slice` (read specific line range) |
+
+### Shell Commands
+| Native Tool | MCP Tool |
+|-------------|----------|
+| `bash` | `manual-slop_run_powershell` |
+
+## Context Amnesia
+You operate statelessly. Each analysis starts fresh.
+Do not assume knowledge from previous analyses or sessions.
+
+## Analysis Start Checklist (MANDATORY)
+
+Before analyzing:
+1. [ ] Read error output/test failure completely
+2. [ ] Identify affected files from traceback
+3. [ ] Use skeleton tools for files >50 lines (`manual-slop_py_get_skeleton`)
+4. [ ] Announce: "Analyzing: [error summary]"
+
+## Analysis Protocol
+
+### 1. Understand the Error
+Read the provided error output, test failure, or log carefully.
+
+### 2. Investigate
+Use MCP tools to understand the context:
+- `manual-slop_read_file` - Read relevant source files
+- `manual-slop_py_find_usages` - Search for related patterns
+- `manual-slop_search_files` - Find related files
+- `manual-slop_get_git_diff` - Check recent changes
+
+### 3. Root Cause Analysis
+Provide a structured analysis:
+
+```
+## Error Analysis
+
+### Summary
+[One-sentence description of the error]
+
+### Root Cause
+[Detailed explanation of why the error occurred]
+
+### Evidence
+[File:line references supporting the analysis]
+
+### Impact
+[What functionality is affected]
+
+### Recommendations
+[Suggested fixes or next steps - but DO NOT implement them]
+```
+
+## Limitations
+- **READ-ONLY**: Do NOT modify any files
+- **ANALYSIS ONLY**: Do NOT implement fixes
+- **NO ASSUMPTIONS**: Base analysis only on provided context and tool output
+
+## Quality Checklist
+- [ ] Analysis is based on actual code/file content
+- [ ] Root cause is specific, not generic
+- [ ] Evidence includes file:line references
+- [ ] Recommendations are actionable but not implemented
+
+## Blocking Protocol
+If you cannot analyze the error:
+1. Start your response with `CANNOT ANALYZE:`
+2. Explain what information is missing
+3. List what would be needed to complete the analysis

.opencode/commands/conductor-implement.md

@@ -0,0 +1,109 @@
+---
+description: Resume or start track implementation following TDD protocol
+agent: tier2-tech-lead
+---
+
+# /conductor-implement
+
+Resume or start implementation of the active track following TDD protocol.
+
+## Prerequisites
+- Run `/conductor-setup` first to load context
+- Ensure a track is active (has `[~]` tasks)
+
+## CRITICAL: Use MCP Tools Only
+
+All research and file operations must use Manual Slop's MCP tools:
+- `manual-slop_py_get_code_outline` - structure analysis
+- `manual-slop_py_get_skeleton` - signatures + docstrings
+- `manual-slop_py_find_usages` - find references
+- `manual-slop_get_git_diff` - recent changes
+- `manual-slop_run_powershell` - shell commands
+
+## Implementation Protocol
+
+1. **Identify Current Task:**
+   - Read active track's `plan.md` via `manual-slop_read_file`
+   - Find the first `[~]` (in-progress) or `[ ]` (pending) task
+   - If phase has no pending tasks, move to next phase
+
+2. **Research Phase (MANDATORY):**
+   Before implementing, use MCP tools to understand context:
+   - `manual-slop_py_get_code_outline` on target files
+   - `manual-slop_py_get_skeleton` on dependencies
+   - `manual-slop_py_find_usages` for related patterns
+   - `manual-slop_get_git_diff` for recent changes
+   - Audit `__init__` methods for existing state
+
+3. **TDD Cycle:**
+
+   ### Red Phase (Write Failing Tests)
+   - Stage current progress: `manual-slop_run_powershell` with `git add .`
+   - Delegate test creation to @tier3-worker:
+     ```
+     @tier3-worker
+     
+     Write tests for: [task description]
+     
+     WHERE: tests/test_file.py:line-range
+     WHAT: Test [specific functionality]
+     HOW: Use pytest, assert [expected behavior]
+     SAFETY: [thread-safety constraints]
+     
+     Use 1-space indentation. Use MCP tools only.
+     ```
+   - Run tests: `manual-slop_run_powershell` with `uv run pytest tests/test_file.py -v`
+   - **CONFIRM TESTS FAIL** - this is the Red phase
+
+   ### Green Phase (Implement to Pass)
+   - Stage current progress: `manual-slop_run_powershell` with `git add .`
+   - Delegate implementation to @tier3-worker:
+     ```
+     @tier3-worker
+     
+     Implement: [task description]
+     
+     WHERE: src/file.py:line-range
+     WHAT: [specific change]
+     HOW: [API calls, patterns to use]
+     SAFETY: [thread-safety constraints]
+     
+     Use 1-space indentation. Use MCP tools only.
+     ```
+   - Run tests: `manual-slop_run_powershell` with `uv run pytest tests/test_file.py -v`
+   - **CONFIRM TESTS PASS** - this is the Green phase
+
+   ### Refactor Phase (Optional)
+   - With passing tests, refactor for clarity
+   - Re-run tests to verify
+
+4. **Commit Protocol (ATOMIC PER-TASK):**
+   Use `manual-slop_run_powershell`:
+   ```powershell
+   git add .
+   git commit -m "feat(scope): description"
+   $hash = git log -1 --format="%H"
+   git notes add -m "Task: [summary]" $hash
+   ```
+   - Update `plan.md`: Change `[~]` to `[x]` with commit SHA
+   - Commit plan update: `git add plan.md && git commit -m "conductor(plan): Mark task complete"`
+
+5. **Repeat for Next Task**
+
+## Error Handling
+If tests fail after Green phase:
+- Delegate analysis to @tier4-qa:
+  ```
+  @tier4-qa
+  
+  Analyze this test failure:
+  
+  [test output]
+  
+  DO NOT fix - provide analysis only. Use MCP tools only.
+  ```
+- Maximum 2 fix attempts before escalating to user
+
+## Phase Completion
+When all tasks in a phase are `[x]`:
+- Run `/conductor-verify` for checkpoint

.opencode/commands/conductor-new-track.md

@@ -0,0 +1,118 @@
+---
+description: Create a new conductor track with spec, plan, and metadata
+agent: tier1-orchestrator
+subtask: true
+---
+
+# /conductor-new-track
+
+Create a new conductor track following the Surgical Methodology.
+
+## Arguments
+$ARGUMENTS - Track name and brief description
+
+## Protocol
+
+1. **Audit Before Specifying (MANDATORY):**
+   Before writing any spec, research the existing codebase:
+   - Use `py_get_code_outline` on relevant files
+   - Use `py_get_definition` on target classes
+   - Use `grep` to find related patterns
+   - Use `get_git_diff` to understand recent changes
+   
+   Document findings in a "Current State Audit" section.
+
+2. **Generate Track ID:**
+   Format: `{name}_{YYYYMMDD}`
+   Example: `async_tool_execution_20260303`
+
+3. **Create Track Directory:**
+   `conductor/tracks/{track_id}/`
+
+4. **Create spec.md:**
+   ```markdown
+   # Track Specification: {Title}
+
+   ## Overview
+   [One-paragraph description]
+
+   ## Current State Audit (as of {commit_sha})
+   ### Already Implemented (DO NOT re-implement)
+   - [Existing feature with file:line reference]
+
+   ### Gaps to Fill (This Track's Scope)
+   - [What's missing that this track will address]
+
+   ## Goals
+   - [Specific, measurable goals]
+
+   ## Functional Requirements
+   - [Detailed requirements]
+
+   ## Non-Functional Requirements
+   - [Performance, security, etc.]
+
+   ## Architecture Reference
+   - docs/guide_architecture.md#section
+   - docs/guide_tools.md#section
+
+   ## Out of Scope
+   - [What this track will NOT do]
+   ```
+
+5. **Create plan.md:**
+   ```markdown
+   # Implementation Plan: {Title}
+
+   ## Phase 1: {Name}
+   Focus: {One-sentence scope}
+
+   - [ ] Task 1.1: {Surgical description with file:line refs}
+   - [ ] Task 1.2: ...
+   - [ ] Task 1.N: Write tests for Phase 1 changes
+   - [ ] Task 1.X: Conductor - User Manual Verification
+
+   ## Phase 2: {Name}
+   ...
+   ```
+
+6. **Create metadata.json:**
+   ```json
+   {
+     "id": "{track_id}",
+     "title": "{title}",
+     "type": "feature|fix|refactor|docs",
+     "status": "planned",
+     "priority": "high|medium|low",
+     "created": "{YYYY-MM-DD}",
+     "depends_on": [],
+     "blocks": []
+   }
+   ```
+
+7. **Update tracks.md:**
+   Add entry to `conductor/tracks.md` registry.
+
+8. **Report:**
+   ```
+   ## Track Created
+
+   **ID:** {track_id}
+   **Location:** conductor/tracks/{track_id}/
+   **Files Created:**
+   - spec.md
+   - plan.md
+   - metadata.json
+
+   **Next Steps:**
+   1. Review spec.md for completeness
+   2. Run `/conductor-implement` to begin execution
+   ```
+
+## Surgical Methodology Checklist
+- [ ] Audited existing code before writing spec
+- [ ] Documented existing implementations with file:line refs
+- [ ] Framed requirements as gaps, not features
+- [ ] Tasks are worker-ready (WHERE/WHAT/HOW/SAFETY)
+- [ ] Referenced architecture docs
+- [ ] Mapped dependencies in metadata

.opencode/commands/conductor-setup.md

@@ -0,0 +1,47 @@
+---
+description: Initialize conductor context — read product docs, verify structure, report readiness
+agent: tier1-orchestrator
+subtask: true
+---
+
+# /conductor-setup
+
+Bootstrap the session with full conductor context. Run this at session start.
+
+## Steps
+
+1. **Read Core Documents:**
+   - `conductor/index.md` — navigation hub
+   - `conductor/product.md` — product vision
+   - `conductor/product-guidelines.md` — UX/code standards
+   - `conductor/tech-stack.md` — technology constraints
+   - `conductor/workflow.md` — task lifecycle (skim; reference during implementation)
+
+2. **Check Active Tracks:**
+   - List all directories in `conductor/tracks/`
+   - Read each `metadata.json` for status
+   - Read each `plan.md` for current task state
+   - Identify the track with `[~]` in-progress tasks
+
+3. **Check Session Context:**
+   - Read `TASKS.md` if it exists — check for IN_PROGRESS or BLOCKED tasks
+   - Read last 3 entries in `JOURNAL.md` for recent activity
+   - Run `git log --oneline -10` for recent commits
+
+4. **Report Readiness:**
+   Present a session startup summary:
+   ```
+   ## Session Ready
+
+   **Active Track:** {track name} — Phase {N}, Task: {current task description}
+   **Recent Activity:** {last journal entry title}
+   **Last Commit:** {git log -1 oneline}
+
+   Ready to:
+   - `/conductor-implement` — resume active track
+   - `/conductor-status` — full status overview
+   - `/conductor-new-track` — start new work
+   ```
+
+## Important
+- This is READ-ONLY — do not modify files

.opencode/commands/conductor-status.md

@@ -0,0 +1,59 @@
+---
+description: Display full status of all conductor tracks and tasks
+agent: tier1-orchestrator
+subtask: true
+---
+
+# /conductor-status
+
+Display comprehensive status of the conductor system.
+
+## Steps
+
+1. **Read Track Index:**
+   - `conductor/tracks.md` — track registry
+   - `conductor/index.md` — navigation hub
+
+2. **Scan All Tracks:**
+   For each track in `conductor/tracks/`:
+   - Read `metadata.json` for status and timestamps
+   - Read `plan.md` for task progress
+   - Count completed vs total tasks
+
+3. **Check TASKS.md:**
+   - List IN_PROGRESS tasks
+   - List BLOCKED tasks
+   - List pending tasks by priority
+
+4. **Recent Activity:**
+   - `git log --oneline -5`
+   - Last 2 entries from `JOURNAL.md`
+
+5. **Report Format:**
+   ```
+   ## Conductor Status
+
+   ### Active Tracks
+   | Track | Status | Progress | Current Task |
+   |-------|--------|----------|--------------|
+   | ... | ... | N/M tasks | ... |
+
+   ### Task Registry (TASKS.md)
+   **In Progress:**
+   - [ ] Task description
+
+   **Blocked:**
+   - [ ] Task description (reason)
+
+   ### Recent Commits
+   - `abc1234` commit message
+
+   ### Recent Journal
+   - YYYY-MM-DD: Entry title
+
+   ### Recommendations
+   - [Next action suggestion]
+   ```
+
+## Important
+- This is READ-ONLY — do not modify files

.opencode/commands/conductor-verify.md

@@ -0,0 +1,92 @@
+---
+description: Verify phase completion and create checkpoint commit
+agent: tier2-tech-lead
+---
+
+# /conductor-verify
+
+Execute phase completion verification and create checkpoint.
+
+## Prerequisites
+- All tasks in the current phase must be marked `[x]`
+- All changes must be committed
+
+## CRITICAL: Use MCP Tools Only
+
+All operations must use Manual Slop's MCP tools:
+- `manual-slop_read_file` - read files
+- `manual-slop_get_git_diff` - check changes
+- `manual-slop_run_powershell` - shell commands
+
+## Verification Protocol
+
+1. **Announce Protocol Start:**
+   Inform user that phase verification has begun.
+
+2. **Determine Phase Scope:**
+   - Find previous phase checkpoint SHA in `plan.md` via `manual-slop_read_file`
+   - If no previous checkpoint, scope is all changes since first commit
+
+3. **List Changed Files:**
+   Use `manual-slop_run_powershell`:
+   ```powershell
+   git diff --name-only <previous_checkpoint_sha> HEAD
+   ```
+
+4. **Verify Test Coverage:**
+   For each code file changed (exclude `.json`, `.md`, `.yaml`):
+   - Check if corresponding test file exists via `manual-slop_search_files`
+   - If missing, create test file via @tier3-worker
+
+5. **Execute Tests in Batches:**
+   **CRITICAL**: Do NOT run full suite. Run max 4 test files at a time.
+   
+   Announce command before execution:
+   ```
+   I will now run: uv run pytest tests/test_file1.py tests/test_file2.py -v
+   ```
+   
+   Use `manual-slop_run_powershell` to execute.
+   
+   If tests fail with large output:
+   - Pipe to log file
+   - Delegate analysis to @tier4-qa
+   - Maximum 2 fix attempts before escalating
+
+6. **Present Results:**
+   ```
+   ## Phase Verification Results
+
+   **Phase:** {phase name}
+   **Files Changed:** {count}
+   **Tests Run:** {count}
+   **Tests Passed:** {count}
+   **Tests Failed:** {count}
+
+   [Detailed results or failure analysis]
+   ```
+
+7. **Await User Confirmation:**
+   **PAUSE** and wait for explicit user approval before proceeding.
+
+8. **Create Checkpoint:**
+   Use `manual-slop_run_powershell`:
+   ```powershell
+   git add .
+   git commit --allow-empty -m "conductor(checkpoint): Phase {N} complete"
+   $hash = git log -1 --format="%H"
+   git notes add -m "Verification: [report summary]" $hash
+   ```
+
+9. **Update Plan:**
+   - Add `[checkpoint: {sha}]` to phase heading in `plan.md`
+   - Use `manual-slop_set_file_slice` or `manual-slop_read_file` + write
+   - Commit: `git add plan.md && git commit -m "conductor(plan): Mark phase complete"`
+
+10. **Announce Completion:**
+    Inform user that phase is complete with checkpoint created.
+
+## Error Handling
+- If any verification fails: HALT and present logs
+- Do NOT proceed without user confirmation
+- Maximum 2 fix attempts per failure

.opencode/commands/mma-tier1-orchestrator.md

@@ -0,0 +1,11 @@
+---
+description: Invoke Tier 1 Orchestrator for product alignment and track initialization
+agent: tier1-orchestrator
+subtask: true
+---
+
+$ARGUMENTS
+
+---
+
+Invoke the Tier 1 Orchestrator with the above context. Focus on product alignment, high-level planning, and track initialization. Follow the Surgical Methodology: audit existing code before specifying, identify gaps not features, and write worker-ready tasks.

.opencode/commands/mma-tier2-tech-lead.md

@@ -0,0 +1,10 @@
+---
+description: Invoke Tier 2 Tech Lead for architectural design and track execution
+agent: tier2-tech-lead
+---
+
+$ARGUMENTS
+
+---
+
+Invoke the Tier 2 Tech Lead with the above context. Follow TDD protocol (Red -> Green -> Refactor), delegate implementation to Tier 3 Workers, and maintain persistent memory throughout track execution. Commit atomically per-task.

.opencode/commands/mma-tier3-worker.md

@@ -0,0 +1,10 @@
+---
+description: Invoke Tier 3 Worker for surgical code implementation
+agent: tier3-worker
+---
+
+$ARGUMENTS
+
+---
+
+Invoke the Tier 3 Worker with the above task. Operate statelessly with context amnesia. Implement the specified change exactly as described. Use 1-space indentation for Python code. Do NOT add comments unless requested.

.opencode/commands/mma-tier4-qa.md

@@ -0,0 +1,10 @@
+---
+description: Invoke Tier 4 QA for error analysis and diagnostics
+agent: tier4-qa
+---
+
+$ARGUMENTS
+
+---
+
+Invoke the Tier 4 QA Agent with the above context. Analyze errors, summarize logs, or verify tests. Provide root cause analysis with file:line evidence. DO NOT implement fixes - analysis only.

AGENTS.md

@@ -0,0 +1,126 @@
+# Manual Slop - OpenCode Configuration
+
+## Project Overview
+
+**Manual Slop** is a local GUI application designed as an experimental, "manual" AI coding assistant. It allows users to curate and send context (files, screenshots, and discussion history) to AI APIs (Gemini and Anthropic). The AI can then execute PowerShell scripts within the project directory to modify files, requiring explicit user confirmation before execution.
+
+## Main Technologies
+
+- **Language:** Python 3.11+
+- **Package Management:** `uv`
+- **GUI Framework:** Dear PyGui (`dearpygui`), ImGui Bundle (`imgui-bundle`)
+- **AI SDKs:** `google-genai` (Gemini), `anthropic`
+- **Configuration:** TOML (`tomli-w`)
+
+## Architecture
+
+- **`gui_legacy.py`:** Main entry point and Dear PyGui application logic
+- **`ai_client.py`:** Unified wrapper for Gemini and Anthropic APIs
+- **`aggregate.py`:** Builds `file_items` context
+- **`mcp_client.py`:** Implements MCP-like tools (26 tools)
+- **`shell_runner.py`:** Sandboxed subprocess wrapper for PowerShell
+- **`project_manager.py`:** Per-project TOML configurations
+- **`session_logger.py`:** Timestamped logging (JSON-L)
+
+## Critical Context (Read First)
+
+- **Tech Stack**: Python 3.11+, Dear PyGui / ImGui, FastAPI, Uvicorn
+- **Main File**: `gui_2.py` (primary GUI), `ai_client.py` (multi-provider LLM abstraction)
+- **Core Mechanic**: GUI orchestrator for LLM-driven coding with 4-tier MMA architecture
+- **Key Integration**: Gemini API, Anthropic API, DeepSeek, Gemini CLI (headless), MCP tools
+- **Platform Support**: Windows (PowerShell)
+- **DO NOT**: Read full files >50 lines without using `py_get_skeleton` or `get_file_summary` first
+
+## Environment
+
+- Shell: PowerShell (pwsh) on Windows
+- Do NOT use bash-specific syntax (use PowerShell equivalents)
+- Use `uv run` for all Python execution
+- Path separators: forward slashes work in PowerShell
+
+## Session Startup Checklist
+
+At the start of each session:
+1. **Check TASKS.md** - look for IN_PROGRESS or BLOCKED tracks
+2. **Review recent JOURNAL.md entries** - scan last 2-3 entries for context
+3. **Run `/conductor-setup`** - load full context
+4. **Run `/conductor-status`** - get overview
+
+## Conductor System
+
+The project uses a spec-driven track system in `conductor/`:
+- **Tracks**: `conductor/tracks/{name}_{YYYYMMDD}/` - spec.md, plan.md, metadata.json
+- **Workflow**: `conductor/workflow.md` - full task lifecycle and TDD protocol
+- **Tech Stack**: `conductor/tech-stack.md` - technology constraints
+- **Product**: `conductor/product.md` - product vision and guidelines
+
+## MMA 4-Tier Architecture
+
+```
+Tier 1: Orchestrator   - product alignment, epic -> tracks
+Tier 2: Tech Lead      - track -> tickets (DAG), architectural oversight
+Tier 3: Worker         - stateless TDD implementation per ticket
+Tier 4: QA             - stateless error analysis, no fixes
+```
+
+## Architecture Fallback
+
+When uncertain about threading, event flow, data structures, or module interactions, consult:
+- **docs/guide_architecture.md**: Thread domains, event system, AI client, HITL mechanism
+- **docs/guide_tools.md**: MCP Bridge security, 26-tool inventory, Hook API endpoints
+- **docs/guide_mma.md**: Ticket/Track data structures, DAG engine, ConductorEngine
+- **docs/guide_simulations.md**: live_gui fixture, Puppeteer pattern, verification
+
+## Development Workflow
+
+1. Run `/conductor-setup` to load session context
+2. Pick active track from `TASKS.md` or `/conductor-status`
+3. Run `/conductor-implement` to resume track execution
+4. Follow TDD: Red (failing tests) -> Green (pass) -> Refactor
+5. Delegate implementation to Tier 3 Workers, errors to Tier 4 QA
+6. On phase completion: run `/conductor-verify` for checkpoint
+
+## Anti-Patterns (Avoid These)
+
+- **Don't read full large files** - use `py_get_skeleton`, `get_file_summary`, `py_get_code_outline` first
+- **Don't implement directly as Tier 2** - delegate to Tier 3 Workers
+- **Don't skip TDD** - write failing tests before implementation
+- **Don't modify tech stack silently** - update `conductor/tech-stack.md` BEFORE implementing
+- **Don't skip phase verification** - run `/conductor-verify` when all tasks in a phase are `[x]`
+- **Don't mix track work** - stay focused on one track at a time
+
+## Code Style
+
+- **IMPORTANT**: DO NOT ADD ***ANY*** COMMENTS unless asked
+- Use 1-space indentation for Python code
+- Use type hints where appropriate
+## Code Style
+
+- **IMPORTANT**: DO NOT ADD ***ANY*** COMMENTS unless asked
+- Use 1-space indentation for Python code
+- Use type hints where appropriate
+- Internal methods/variables prefixed with underscore
+
+### CRITICAL: Native Edit Tool Destroys Indentation
+
+The native `Edit` tool DESTROYS 1-space indentation and converts to 4-space.
+
+**NEVER use native `edit` tool on Python files.**
+
+Instead, use Manual Slop MCP tools:
+- `manual-slop_py_update_definition` - Replace function/class
+- `manual-slop_set_file_slice` - Replace line range
+- `manual-slop_py_set_signature` - Replace signature only
+
+Or use Python subprocess with `newline=''` to preserve line endings:
+```python
+python -c "
+with open('file.py', 'r', encoding='utf-8', newline='') as f:
+    content = f.read()
+content = content.replace(old, new)
+with open('file.py', 'w', encoding='utf-8', newline='') as f:
+    f.write(content)
+"
+```
+
+## Quality Gates

ARCHITECTURE.md

@@ -0,0 +1,58 @@
+# ARCHITECTURE.md
+
+## Tech Stack
+- **Framework**: [Primary framework/language]
+- **Database**: [Database system]
+- **Frontend**: [Frontend technology]
+- **Backend**: [Backend technology]
+- **Infrastructure**: [Hosting/deployment]
+- **Build Tools**: [Build system]
+
+## Directory Structure
+```
+project/
+├── src/              # Source code
+├── tests/            # Test files
+├── docs/             # Documentation
+├── config/           # Configuration files
+└── scripts/          # Build/deployment scripts
+```
+
+## Key Architectural Decisions
+
+### [Decision 1]
+**Context**: [Why this decision was needed]
+**Decision**: [What was decided]
+**Rationale**: [Why this approach was chosen]
+**Consequences**: [Trade-offs and implications]
+
+## Component Architecture
+
+### [ComponentName] Structure <!-- #component-anchor -->
+```typescript
+// Major classes with exact line numbers
+class MainClass { /* lines 100-500 */ }    // <!-- #main-class -->
+class Helper { /* lines 501-600 */ }       // <!-- #helper-class -->
+```
+
+## System Flow Diagram
+```
+[User] -> [Frontend] -> [API] -> [Database]
+           |            |
+           v            v
+       [Cache]     [External Service]
+```
+
+## Common Patterns
+
+### [Pattern Name]
+**When to use**: [Circumstances]
+**Implementation**: [How to implement]
+**Example**: [Code example with line numbers]
+
+## Keywords <!-- #keywords -->
+- architecture
+- system design
+- tech stack
+- components
+- patterns

BUILD.md

@@ -0,0 +1,103 @@
+# BUILD.md
+
+## Prerequisites
+- [Runtime requirements]
+- [Development tools needed]
+- [Environment setup]
+
+## Build Commands
+
+### Development
+```bash
+# Start development server
+npm run dev
+
+# Run in watch mode
+npm run watch
+```
+
+### Production
+```bash
+# Build for production
+npm run build
+
+# Start production server
+npm start
+```
+
+### Testing
+```bash
+# Run all tests
+npm test
+
+# Run tests in watch mode
+npm run test:watch
+
+# Run specific test file
+npm test -- filename
+```
+
+### Linting & Formatting
+```bash
+# Lint code
+npm run lint
+
+# Fix linting issues
+npm run lint:fix
+
+# Format code
+npm run format
+```
+
+## CI/CD Pipeline
+
+### GitHub Actions
+```yaml
+# .github/workflows/main.yml
+name: CI/CD
+on: [push, pull_request]
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v3
+      - name: Setup Node.js
+        uses: actions/setup-node@v3
+        with:
+          node-version: '18'
+      - run: npm ci
+      - run: npm test
+      - run: npm run build
+```
+
+## Deployment
+
+### Staging
+1. [Deployment steps]
+2. [Verification steps]
+
+### Production
+1. [Pre-deployment checklist]
+2. [Deployment steps]
+3. [Post-deployment verification]
+
+## Rollback Procedures
+1. [Emergency rollback steps]
+2. [Database rollback if needed]
+3. [Verification steps]
+
+## Troubleshooting
+
+### Common Issues
+**Issue**: [Problem description]
+**Solution**: [How to fix]
+
+### Build Failures
+- [Common build errors and solutions]
+
+## Keywords <!-- #keywords -->
+- build
+- deployment
+- ci/cd
+- testing
+- production

CLAUDE.md

@@ -0,0 +1,118 @@
+# CLAUDE.md
+<!-- Generated by Claude Conductor v2.0.0 -->
+
+This file provides guidance to Claude Code when working with this repository.
+
+## Critical Context (Read First)
+- **Tech Stack**: Python 3.11+, Dear PyGui / ImGui, FastAPI, Uvicorn
+- **Main File**: `gui_2.py` (primary GUI), `ai_client.py` (multi-provider LLM abstraction)
+- **Core Mechanic**: GUI orchestrator for LLM-driven coding with 4-tier MMA architecture
+- **Key Integration**: Gemini API, Anthropic API, DeepSeek, Gemini CLI (headless), MCP tools
+- **Platform Support**: Windows (PowerShell) — single developer, local use
+- **DO NOT**: Read full files >50 lines without using `py_get_skeleton` or `get_file_summary` first. Do NOT perform heavy implementation directly — delegate to Tier 3 Workers.
+
+## Environment
+- Shell: PowerShell (pwsh) on Windows
+- Do NOT use bash-specific syntax (use PowerShell equivalents)
+- Use `uv run` for all Python execution
+- Path separators: forward slashes work in PowerShell
+- **Shell execution in Claude Code**: The `Bash` tool runs in a mingw sandbox on Windows and produces unreliable/empty output. Use `run_powershell` MCP tool for ALL shell commands (git, tests, scans). Bash is last-resort only when MCP server is not running.
+
+## Session Startup Checklist
+**IMPORTANT**: At the start of each session:
+1. **Check TASKS.md** — look for IN_PROGRESS or BLOCKED tracks
+2. **Review recent JOURNAL.md entries** — scan last 2-3 entries for context
+3. **If resuming work**: run `/conductor-setup` to load full context
+4. **If starting fresh**: run `/conductor-status` for overview
+
+## Quick Reference
+**GUI Entry**: `gui_2.py` — Primary ImGui interface
+**AI Client**: `ai_client.py` — Multi-provider abstraction (Gemini, Anthropic, DeepSeek)
+**MCP Client**: `mcp_client.py:773-831` — Tool dispatch (26 tools)
+**Project Manager**: `project_manager.py` — Context & file management
+**MMA Engine**: `multi_agent_conductor.py:15-100` — ConductorEngine orchestration
+**Tech Lead**: `conductor_tech_lead.py` — Tier 2 ticket generation
+**DAG Engine**: `dag_engine.py` — Task dependency resolution
+**Session Logger**: `session_logger.py` — Audit trails (JSON-L + markdown)
+**Shell Runner**: `shell_runner.py` — PowerShell execution (60s timeout)
+**Models**: `models.py:6-84` — Ticket and Track data structures
+**File Cache**: `file_cache.py` — ASTParser with tree-sitter skeletons
+**Summarizer**: `summarize.py` — Heuristic file summaries
+**Outliner**: `outline_tool.py` — Code outline with line ranges
+
+## Conductor System
+The project uses a spec-driven track system in `conductor/`:
+- **Tracks**: `conductor/tracks/{name}_{YYYYMMDD}/` — spec.md, plan.md, metadata.json
+- **Workflow**: `conductor/workflow.md` — full task lifecycle and TDD protocol
+- **Tech Stack**: `conductor/tech-stack.md` — technology constraints
+- **Product**: `conductor/product.md` — product vision and guidelines
+
+### Conductor Commands (Claude Code slash commands)
+- `/conductor-setup` — bootstrap session with conductor context
+- `/conductor-status` — show all track status
+- `/conductor-new-track` — create a new track (Tier 1)
+- `/conductor-implement` — execute a track (Tier 2 — delegates to Tier 3/4)
+- `/conductor-verify` — phase completion verification and checkpointing
+
+### MMA Tier Commands
+- `/mma-tier1-orchestrator` — product alignment, planning
+- `/mma-tier2-tech-lead` — track execution, architectural oversight
+- `/mma-tier3-worker` — stateless TDD implementation
+- `/mma-tier4-qa` — stateless error analysis
+
+### Delegation (Tier 2 spawns Tier 3/4)
+```powershell
+uv run python scripts\claude_mma_exec.py --role tier3-worker "Task prompt here"
+uv run python scripts\claude_mma_exec.py --role tier4-qa "Error analysis prompt"
+```
+
+## Current State
+- [x] Multi-provider AI client (Gemini, Anthropic, DeepSeek)
+- [x] Dear PyGui / ImGui GUI with multi-panel interface
+- [x] MMA 4-tier orchestration engine
+- [x] Custom MCP tools (26 tools via mcp_client.py)
+- [x] Session logging and audit trails
+- [x] Gemini CLI headless adapter
+- [x] Claude Code conductor integration
+- [~] AI-Optimized Python Style Refactor (Phase 3 — type hints for UI modules)
+- [~] Robust Live Simulation Verification (Phase 2 — Epic/Track verification)
+- [ ] Documentation Refresh and Context Cleanup
+
+## Development Workflow
+1. Run `/conductor-setup` to load session context
+2. Pick active track from `TASKS.md` or `/conductor-status`
+3. Run `/conductor-implement` to resume track execution
+4. Follow TDD: Red (failing tests) → Green (pass) → Refactor
+5. Delegate implementation to Tier 3 Workers, errors to Tier 4 QA
+6. On phase completion: run `/conductor-verify` for checkpoint
+
+## Anti-Patterns (Avoid These)
+- **Don't read full large files** — use `py_get_skeleton`, `get_file_summary`, `py_get_code_outline` first (Research-First Protocol)
+- **Don't implement directly as Tier 2** — delegate to Tier 3 Workers via `claude_mma_exec.py`
+- **Don't skip TDD** — write failing tests before implementation
+- **Don't modify tech stack silently** — update `conductor/tech-stack.md` BEFORE implementing
+- **Don't skip phase verification** — run `/conductor-verify` when all tasks in a phase are `[x]`
+- **Don't mix track work** — stay focused on one track at a time
+
+## MCP Tools (available via manual-slop MCP server)
+When the MCP server is running, these tools are available natively:
+`py_get_skeleton`, `py_get_code_outline`, `py_get_definition`, `py_update_definition`,
+`py_get_signature`, `py_set_signature`, `py_get_class_summary`, `py_find_usages`,
+`py_get_imports`, `py_check_syntax`, `py_get_hierarchy`, `py_get_docstring`,
+`get_file_summary`, `get_file_slice`, `set_file_slice`, `get_git_diff`, `get_tree`,
+`search_files`, `read_file`, `list_directory`, `web_search`, `fetch_url`,
+`run_powershell`, `get_ui_performance`, `py_get_var_declaration`, `py_set_var_declaration`
+
+## Journal Update Requirements
+Update JOURNAL.md after:
+- Completing any significant feature or fix
+- Encountering and resolving errors
+- End of each work session
+- Making architectural decisions
+Format: What/Why/How/Issues/Result structure
+
+## Task Management Integration
+- **TASKS.md**: Quick-read pointer to active conductor tracks
+- **conductor/tracks/*/plan.md**: Detailed task state (source of truth)
+- **JOURNAL.md**: Completed work history with `|TASK:ID|` tags
+- **ERRORS.md**: P0/P1 error tracking

CONDUCTOR.md

@@ -0,0 +1,511 @@
+# CONDUCTOR.md
+<!-- Generated by Claude Conductor v2.0.0 -->
+
+> _Read me first. Every other doc is linked below._
+
+## Critical Context (Read First)
+- **Tech Stack**: [List core technologies]
+- **Main File**: [Primary code file and line count]
+- **Core Mechanic**: [One-line description]
+- **Key Integration**: [Important external services]
+- **Platform Support**: [Deployment targets]
+- **DO NOT**: [Critical things to avoid]
+
+## Table of Contents
+1. [Architecture](ARCHITECTURE.md) - Tech stack, folder structure, infrastructure
+2. [Design Tokens](DESIGN.md) - Colors, typography, visual system
+3. [UI/UX Patterns](UIUX.md) - Components, interactions, accessibility
+4. [Runtime Config](CONFIG.md) - Environment variables, feature flags
+5. [Data Model](DATA_MODEL.md) - Database schema, entities, relationships
+6. [API Contracts](API.md) - Endpoints, request/response formats, auth
+7. [Build & Release](BUILD.md) - Build process, deployment, CI/CD
+8. [Testing Guide](TEST.md) - Test strategies, E2E scenarios, coverage
+9. [Operational Playbooks](PLAYBOOKS/DEPLOY.md) - Deployment, rollback, monitoring
+10. [Contributing](CONTRIBUTING.md) - Code style, PR process, conventions
+11. [Error Ledger](ERRORS.md) - Critical P0/P1 error tracking
+12. [Task Management](TASKS.md) - Active tasks, phase tracking, context preservation
+
+## Quick Reference
+**Main Constants**: `[file:lines]` - Description  
+**Core Class**: `[file:lines]` - Description  
+**Key Function**: `[file:lines]` - Description  
+[Include 10-15 most accessed code locations]
+
+## Current State
+- [x] Feature complete
+- [ ] Feature in progress
+- [ ] Feature planned
+[Track active work]
+
+## Development Workflow
+[5-6 steps for common workflow]
+
+## Task Templates
+### 1. [Common Task Name]
+1. Step with file:line reference
+2. Step with specific action
+3. Test step
+4. Documentation update
+
+[Include 3-5 templates]
+
+## Anti-Patterns (Avoid These)
+❌ **Don't [action]** - [Reason]  
+[List 5-6 critical mistakes]
+
+## Version History
+- **v1.0.0** - Initial release
+- **v1.1.0** - Feature added (see JOURNAL.md YYYY-MM-DD)  
+[Link major versions to journal entries]
+
+## Continuous Engineering Journal <!-- do not remove -->
+
+Claude, keep an ever-growing changelog in [`JOURNAL.md`](JOURNAL.md).
+
+### What to Journal
+- **Major changes**: New features, significant refactors, API changes
+- **Bug fixes**: What broke, why, and how it was fixed
+- **Frustration points**: Problems that took multiple attempts to solve
+- **Design decisions**: Why we chose one approach over another
+- **Performance improvements**: Before/after metrics
+- **User feedback**: Notable issues or requests
+- **Learning moments**: New techniques or patterns discovered
+
+### Journal Format
+\```
+## YYYY-MM-DD HH:MM
+
+### [Short Title]
+- **What**: Brief description of the change
+- **Why**: Reason for the change
+- **How**: Technical approach taken
+- **Issues**: Any problems encountered
+- **Result**: Outcome and any metrics
+
+### [Short Title] |ERROR:ERR-YYYY-MM-DD-001|
+- **What**: Critical P0/P1 error description
+- **Why**: Root cause analysis
+- **How**: Fix implementation
+- **Issues**: Debugging challenges
+- **Result**: Resolution and prevention measures
+
+### [Task Title] |TASK:TASK-YYYY-MM-DD-001|
+- **What**: Task implementation summary
+- **Why**: Part of [Phase Name] phase
+- **How**: Technical approach and key decisions
+- **Issues**: Blockers encountered and resolved
+- **Result**: Task completed, findings documented in ARCHITECTURE.md
+\```
+
+### Compaction Rule
+When `JOURNAL.md` exceeds **500 lines**:
+1. Claude summarizes the oldest half into `JOURNAL_ARCHIVE/<year>-<month>.md`
+2. Remaining entries stay in `JOURNAL.md` so the file never grows unbounded
+
+> ⚠️ Claude must NEVER delete raw history—only move & summarize.
+
+### 2. ARCHITECTURE.md
+**Purpose**: System design, tech stack decisions, and code structure with line numbers.
+
+**Required Elements**:
+- Technology stack listing
+- Directory structure diagram
+- Key architectural decisions with rationale
+- Component architecture with exact line numbers
+- System flow diagram (ASCII art)
+- Common patterns section
+- Keywords for search optimization
+
+**Line Number Format**:
+\```
+#### ComponentName Structure <!-- #component-anchor -->
+\```typescript
+// Major classes with exact line numbers
+class MainClass { /* lines 100-500 */ }    // <!-- #main-class -->
+class Helper { /* lines 501-600 */ }       // <!-- #helper-class -->
+\```
+\```
+
+### 3. DESIGN.md
+**Purpose**: Visual design system, styling, and theming documentation.
+
+**Required Sections**:
+- Typography system
+- Color palette (with hex values)
+- Visual effects specifications
+- Character/entity design
+- UI/UX component styling
+- Animation system
+- Mobile design considerations
+- Accessibility guidelines
+- Keywords section
+
+### 4. DATA_MODEL.md
+**Purpose**: Database schema, application models, and data structures.
+
+**Required Elements**:
+- Database schema (SQL)
+- Application data models (TypeScript/language interfaces)
+- Validation rules
+- Common queries
+- Data migration history
+- Keywords for entities
+
+### 5. API.md
+**Purpose**: Complete API documentation with examples.
+
+**Structure for Each Endpoint**:
+\```
+### Endpoint Name
+
+\```http
+METHOD /api/endpoint
+\```
+
+#### Request
+\```json
+{
+  "field": "type"
+}
+\```
+
+#### Response
+\```json
+{
+  "field": "value"
+}
+\```
+
+#### Details
+- **Rate limit**: X requests per Y seconds
+- **Auth**: Required/Optional
+- **Notes**: Special considerations
+\```
+
+### 6. CONFIG.md
+**Purpose**: Runtime configuration, environment variables, and settings.
+
+**Required Sections**:
+- Environment variables (required and optional)
+- Application configuration constants
+- Feature flags
+- Performance tuning settings
+- Security configuration
+- Common patterns for configuration changes
+
+### 7. BUILD.md
+**Purpose**: Build process, deployment, and CI/CD documentation.
+
+**Include**:
+- Prerequisites
+- Build commands
+- CI/CD pipeline configuration
+- Deployment steps
+- Rollback procedures
+- Troubleshooting guide
+
+### 8. TEST.md
+**Purpose**: Testing strategies, patterns, and examples.
+
+**Sections**:
+- Test stack and tools
+- Running tests commands
+- Test structure
+- Coverage goals
+- Common test patterns
+- Debugging tests
+
+### 9. UIUX.md
+**Purpose**: Interaction patterns, user flows, and behavior specifications.
+
+**Cover**:
+- Input methods
+- State transitions
+- Component behaviors
+- User flows
+- Accessibility patterns
+- Performance considerations
+
+### 10. CONTRIBUTING.md
+**Purpose**: Guidelines for contributors.
+
+**Include**:
+- Code of conduct
+- Development setup
+- Code style guide
+- Commit message format
+- PR process
+- Common patterns
+
+### 11. PLAYBOOKS/DEPLOY.md
+**Purpose**: Step-by-step operational procedures.
+
+**Format**:
+- Pre-deployment checklist
+- Deployment steps (multiple options)
+- Post-deployment verification
+- Rollback procedures
+- Troubleshooting
+
+### 12. ERRORS.md (Critical Error Ledger)
+**Purpose**: Track and resolve P0/P1 critical errors with full traceability.
+
+**Required Structure**:
+\```
+# Critical Error Ledger <!-- auto-maintained -->
+
+## Schema
+| ID | First seen | Status | Severity | Affected area | Link to fix |
+|----|------------|--------|----------|---------------|-------------|
+
+## Active Errors
+[New errors added here, newest first]
+
+## Resolved Errors
+[Moved here when fixed, with links to fixes]
+\```
+
+**Error ID Format**: `ERR-YYYY-MM-DD-001` (increment for multiple per day)
+
+**Severity Definitions**:
+- **P0**: Complete outage, data loss, security breach
+- **P1**: Major functionality broken, significant performance degradation
+- **P2**: Minor functionality (not tracked in ERRORS.md)
+- **P3**: Cosmetic issues (not tracked in ERRORS.md)
+
+**Claude's Error Logging Process**:
+1. When P0/P1 error occurs, immediately add to Active Errors
+2. Create corresponding JOURNAL.md entry with details
+3. When resolved:
+   - Move to Resolved Errors section
+   - Update status to "resolved"
+   - Add commit hash and PR link
+   - Add `|ERROR:<ID>|` tag to JOURNAL.md entry
+   - Link back to JOURNAL entry from ERRORS.md
+
+### 13. TASKS.md (Active Task Management)
+**Purpose**: Track ongoing work with phase awareness and context preservation between sessions.
+
+**IMPORTANT**: TASKS.md complements Claude's built-in todo system - it does NOT replace it:
+- Claude's todos: For immediate task tracking within a session
+- TASKS.md: For preserving context and state between sessions
+
+**Required Structure**:
+```
+# Task Management
+
+## Active Phase
+**Phase**: [High-level project phase name]
+**Started**: YYYY-MM-DD
+**Target**: YYYY-MM-DD
+**Progress**: X/Y tasks completed
+
+## Current Task
+**Task ID**: TASK-YYYY-MM-DD-NNN
+**Title**: [Descriptive task name]
+**Status**: PLANNING | IN_PROGRESS | BLOCKED | TESTING | COMPLETE
+**Started**: YYYY-MM-DD HH:MM
+**Dependencies**: [List task IDs this depends on]
+
+### Task Context
+<!-- Critical information needed to resume this task -->
+- **Previous Work**: [Link to related tasks/PRs]
+- **Key Files**: [Primary files being modified with line ranges]
+- **Environment**: [Specific config/versions if relevant]
+- **Next Steps**: [Immediate actions when resuming]
+
+### Findings & Decisions
+- **FINDING-001**: [Discovery that affects approach]
+- **DECISION-001**: [Technical choice made] → Link to ARCHITECTURE.md
+- **BLOCKER-001**: [Issue preventing progress] → Link to resolution
+
+### Task Chain
+1. ✅ [Completed prerequisite task] (TASK-YYYY-MM-DD-001)
+2. 🔄 [Current task] (CURRENT)
+3. ⏳ [Next planned task]
+4. ⏳ [Future task in phase]
+```
+
+**Task Management Rules**:
+1. **One Active Task**: Only one task should be IN_PROGRESS at a time
+2. **Context Capture**: Before switching tasks, capture all context needed to resume
+3. **Findings Documentation**: Record unexpected discoveries that impact the approach
+4. **Decision Linking**: Link architectural decisions to ARCHITECTURE.md
+5. **Completion Trigger**: When task completes:
+   - Generate JOURNAL.md entry with task summary
+   - Archive task details to TASKS_ARCHIVE/YYYY-MM/TASK-ID.md
+   - Load next task from chain or prompt for new phase
+
+**Task States**:
+- **PLANNING**: Defining approach and breaking down work
+- **IN_PROGRESS**: Actively working on implementation
+- **BLOCKED**: Waiting on external dependency or decision
+- **TESTING**: Implementation complete, validating functionality
+- **COMPLETE**: Task finished and documented
+
+**Integration with Journal**:
+- Each completed task auto-generates a journal entry
+- Journal references task ID for full context
+- Critical findings promoted to relevant documentation
+
+## Documentation Optimization Rules
+
+### 1. Line Number Anchors
+- Add exact line numbers for every class, function, and major code section
+- Format: `**Class Name (Lines 100-200)**`
+- Add HTML anchors: `<!-- #class-name -->`
+- Update when code structure changes significantly
+
+### 2. Quick Reference Card
+- Place in CLAUDE.md after Table of Contents
+- Include 10-15 most common code locations
+- Format: `**Feature**: `file:lines` - Description`
+
+### 3. Current State Tracking
+- Use checkbox format in CLAUDE.md
+- `- [x] Completed feature`
+- `- [ ] In-progress feature`
+- Update after each work session
+
+### 4. Task Templates
+- Provide 3-5 step-by-step workflows
+- Include specific line numbers
+- Reference files that need updating
+- Add test/verification steps
+
+### 5. Keywords Sections
+- Add to each major .md file
+- List alternative search terms
+- Format: `## Keywords <!-- #keywords -->`
+- Include synonyms and related terms
+
+### 6. Anti-Patterns
+- Use ❌ emoji for clarity
+- Explain why each is problematic
+- Include 5-6 critical mistakes
+- Place prominently in CLAUDE.md
+
+### 7. System Flow Diagrams
+- Use ASCII art for simplicity
+- Show data/control flow
+- Keep visual and readable
+- Place in ARCHITECTURE.md
+
+### 8. Common Patterns
+- Add to relevant docs (CONFIG.md, ARCHITECTURE.md)
+- Show exact code changes needed
+- Include before/after examples
+- Reference specific functions
+
+### 9. Version History
+- Link to JOURNAL.md entries
+- Format: `v1.0.0 - Feature (see JOURNAL.md YYYY-MM-DD)`
+- Track major changes only
+
+### 10. Cross-Linking
+- Link between related sections
+- Use relative paths: `[Link](./FILE.md#section)`
+- Ensure bidirectional linking where appropriate
+
+## Journal System Setup
+
+### JOURNAL.md Structure
+\```
+# Engineering Journal
+
+## YYYY-MM-DD HH:MM
+
+### [Descriptive Title]
+- **What**: Brief description of the change
+- **Why**: Reason for the change
+- **How**: Technical approach taken
+- **Issues**: Any problems encountered
+- **Result**: Outcome and any metrics
+
+---
+
+[Entries continue chronologically]
+\```
+
+### Journal Best Practices
+1. **Entry Timing**: Add entry immediately after significant work
+2. **Detail Level**: Include enough detail to understand the change months later
+3. **Problem Documentation**: Especially document multi-attempt solutions
+4. **Learning Moments**: Capture new techniques discovered
+5. **Metrics**: Include performance improvements, time saved, etc.
+
+### Archive Process
+When JOURNAL.md exceeds 500 lines:
+1. Create `JOURNAL_ARCHIVE/` directory
+2. Move oldest 250 lines to `JOURNAL_ARCHIVE/YYYY-MM.md`
+3. Add summary header to archive file
+4. Keep recent entries in main JOURNAL.md
+
+## Implementation Steps
+
+### Phase 1: Initial Setup (30-60 minutes)
+1. **Create CLAUDE.md** with all required sections
+2. **Fill Critical Context** with 6 essential facts
+3. **Create Table of Contents** with placeholder links
+4. **Add Quick Reference** with top 10-15 code locations
+5. **Set up Journal section** with formatting rules
+
+### Phase 2: Core Documentation (2-4 hours)
+1. **Create each .md file** from the list above
+2. **Add Keywords section** to each file
+3. **Cross-link between files** where relevant
+4. **Add line numbers** to code references
+5. **Create PLAYBOOKS/ directory** with DEPLOY.md
+6. **Create ERRORS.md** with schema table
+
+### Phase 3: Optimization (1-2 hours)
+1. **Add Task Templates** to CLAUDE.md
+2. **Create ASCII system flow** in ARCHITECTURE.md
+3. **Add Common Patterns** sections
+4. **Document Anti-Patterns**
+5. **Set up Version History**
+
+### Phase 4: First Journal Entry
+Create initial JOURNAL.md entry documenting the setup:
+\```
+## YYYY-MM-DD HH:MM
+
+### Documentation Framework Implementation
+- **What**: Implemented CLAUDE.md modular documentation system
+- **Why**: Improve AI navigation and code maintainability
+- **How**: Split monolithic docs into focused modules with cross-linking
+- **Issues**: None - clean implementation
+- **Result**: [Number] documentation files created with full cross-referencing
+\```
+
+## Maintenance Guidelines
+
+### Daily
+- Update JOURNAL.md with significant changes
+- Mark completed items in Current State
+- Update line numbers if major refactoring
+
+### Weekly
+- Review and update Quick Reference section
+- Check for broken cross-links
+- Update Task Templates if workflows change
+
+### Monthly
+- Review Keywords sections for completeness
+- Update Version History
+- Check if JOURNAL.md needs archiving
+
+### Per Release
+- Update Version History in CLAUDE.md
+- Create comprehensive JOURNAL.md entry
+- Review all documentation for accuracy
+- Update Current State checklist
+
+## Benefits of This System
+
+1. **AI Efficiency**: Claude can quickly navigate to exact code locations
+2. **Modularity**: Easy to update specific documentation without affecting others
+3. **Discoverability**: New developers/AI can quickly understand the project
+4. **History Tracking**: Complete record of changes and decisions
+5. **Task Automation**: Templates reduce repetitive instructions
+6. **Error Prevention**: Anti-patterns prevent common mistakes

Dockerfile

@@ -0,0 +1,34 @@
+# Use python:3.11-slim as a base
+FROM python:3.11-slim
+
+# Set environment variables
+# UV_SYSTEM_PYTHON=1 allows uv to install into the system site-packages
+ENV PYTHONDONTWRITEBYTECODE=1 
+    PYTHONUNBUFFERED=1 
+    UV_SYSTEM_PYTHON=1
+
+# Install system dependencies and uv
+RUN apt-get update && apt-get install -y --no-install-recommends 
+    curl 
+    ca-certificates 
+    && rm -rf /var/lib/apt/lists/* 
+    && curl -LsSf https://astral.sh/uv/install.sh | sh 
+    && mv /root/.local/bin/uv /usr/local/bin/uv
+
+# Set the working directory in the container
+WORKDIR /app
+
+# Copy dependency files first to leverage Docker layer caching
+COPY pyproject.toml requirements.txt* ./
+
+# Install dependencies via uv
+RUN if [ -f requirements.txt ]; then uv pip install --no-cache -r requirements.txt; fi
+
+# Copy the rest of the application code
+COPY . .
+
+# Expose port 8000 for the headless API/service
+EXPOSE 8000
+
+# Set the entrypoint to run the app in headless mode
+ENTRYPOINT ["python", "gui_2.py", "--headless"]

GEMINI.md

@@ -0,0 +1,47 @@
+# Project Overview
+
+**Manual Slop** is a local GUI application designed as an experimental, "manual" AI coding assistant. It allows users to curate and send context (files, screenshots, and discussion history) to AI APIs (Gemini and Anthropic). The AI can then execute PowerShell scripts within the project directory to modify files, requiring explicit user confirmation before execution.
+
+**Main Technologies:**
+*   **Language:** Python 3.11+
+*   **Package Management:** `uv`
+*   **GUI Framework:** Dear PyGui (`dearpygui`), ImGui Bundle (`imgui-bundle`)
+*   **AI SDKs:** `google-genai` (Gemini), `anthropic`
+*   **Configuration:** TOML (`tomli-w`)
+
+**Architecture:**
+*   **`gui_legacy.py`:** The main entry point and Dear PyGui application logic. Handles all panels, layouts, user input, and confirmation dialogs.
+*   **`ai_client.py`:** A unified wrapper for both Gemini and Anthropic APIs. Manages sessions, tool/function-call loops, token estimation, and context history management.
+*   **`aggregate.py`:** Responsible for building the `file_items` context. It reads project configurations, collects files and screenshots, and builds the context into markdown format to send to the AI.
+*   **`mcp_client.py`:** Implements MCP-like tools (e.g., `read_file`, `list_directory`, `search_files`, `web_search`) as native functions that the AI can call. Enforces a strict allowlist for file access.
+*   **`shell_runner.py`:** A sandboxed subprocess wrapper that executes PowerShell scripts (`powershell -NoProfile -NonInteractive -Command`) provided by the AI.
+*   **`project_manager.py`:** Manages per-project TOML configurations (`manual_slop.toml`), serializes discussion entries, and integrates with git (e.g., fetching current commit).
+*   **`session_logger.py`:** Handles timestamped logging of communication history (JSON-L) and tool calls (saving generated `.ps1` files).
+
+# Building and Running
+
+*   **Setup:** The application uses `uv` for dependency management. Ensure `uv` is installed.
+*   **Credentials:** You must create a `credentials.toml` file in the root directory to store your API keys:
+    ```toml
+    [gemini]
+    api_key = "****"
+    [anthropic]
+    api_key = "****"
+    ```
+*   **Run the Application:**
+    ```powershell
+    uv run .\gui_2.py
+    ```
+
+# Development Conventions
+
+*   **Configuration Management:** The application uses two tiers of configuration:
+    *   `config.toml`: Global settings (UI theme, active provider, list of project paths).
+    *   `manual_slop.toml`: Per-project settings (files to track, discussion history, specific system prompts).
+*   **Tool Execution:** The AI acts primarily by generating PowerShell scripts. These scripts MUST be confirmed by the user via a GUI modal before execution. The AI also has access to read-only MCP-style file exploration tools and web search capabilities.
+*   **Context Refresh:** After every tool call that modifies the file system, the application automatically refreshes the file contents in the context using the files' `mtime` to optimize reads.
+*   **UI State Persistence:** Window layouts and docking arrangements are automatically saved to and loaded from `dpg_layout.ini`.
+*   **Code Style:**
+    *   Use type hints where appropriate.
+    *   Internal methods and variables are generally prefixed with an underscore (e.g., `_flush_to_project`, `_do_generate`).
+*   **Logging:** All API communications are logged to `logs/comms_<ts>.log`. All executed scripts are saved to `scripts/generated/`.

JOURNAL.md

@@ -0,0 +1,108 @@
+# Engineering Journal
+
+## 2026-02-28 14:43
+
+### Documentation Framework Implementation
+- **What**: Implemented Claude Conductor modular documentation system
+- **Why**: Improve AI navigation and code maintainability
+- **How**: Used `npx claude-conductor` to initialize framework
+- **Issues**: None - clean implementation
+- **Result**: Documentation framework successfully initialized
+
+---
+
+---
+
+## 2026-03-02
+
+### Track: context_token_viz_20260301 — Completed |TASK:context_token_viz_20260301|
+- **What**: Token budget visualization panel (all 3 phases)
+- **Why**: Zero visibility into context window usage; `get_history_bleed_stats` existed but had no UI
+- **How**: Extended `get_history_bleed_stats` with `_add_bleed_derived` helper (adds 8 derived fields); added `_render_token_budget_panel` with color-coded progress bar, breakdown table, trim warning, Gemini/Anthropic cache status; 3 auto-refresh triggers (`_token_stats_dirty` flag); `/api/gui/token_stats` endpoint; `--timeout` flag on `claude_mma_exec.py`
+- **Issues**: `set_file_slice` dropped `def _render_message_panel` line — caught by outline check, fixed with 1-line insert. Tier 3 delegation via `run_powershell` hard-capped at 60s — implemented changes directly per user approval; added `--timeout` flag for future use.
+- **Result**: 17 passing tests, all phases verified by user. Token panel visible in AI Settings under "Token Budget". Commits: 5bfb20f → d577457.
+
+### Next: mma_agent_focus_ux (planned, not yet tracked)
+- **What**: Per-agent filtering for MMA observability panels (comms, tool calls, discussion, token budget)
+- **Why**: All panels are global/session-scoped; in MMA mode with 4 tiers, data from all agents mixes. No way to isolate what a specific tier is doing.
+- **Gap**: `_comms_log` and `_tool_log` have no tier/agent tag. `mma_streams` stream_id is the only per-agent key that exists.
+- **See**: TASKS.md for full audit and implementation intent.
+
+---
+
+## 2026-03-02 (Session 2)
+
+### Tracks Initialized: feature_bleed_cleanup + mma_agent_focus_ux |TASK:feature_bleed_cleanup_20260302| |TASK:mma_agent_focus_ux_20260302|
+- **What**: Audited codebase for feature bleed; initialized 2 new conductor tracks
+- **Why**: Entropy from Tier 2 track implementations — redundant code, dead methods, layout regressions, no tier context in observability
+- **Bleed findings** (gui_2.py): Dead duplicate `_render_comms_history_panel` (3041-3073, stale `type` key, wrong method ref); dead `begin_main_menu_bar()` block (1680-1705, Quit has never worked); 4 duplicate `__init__` assignments; double "Token Budget" label with no collapsing header
+- **Agent focus findings** (ai_client.py + conductors): No `current_tier` var; Tier 3 swaps callback but never stamps tier; Tier 2 doesn't swap at all; `_tool_log` is untagged tuple list
+- **Result**: 2 tracks committed (4f11d1e, c1a86e2). Bleed cleanup is active; agent focus depends on it.
+
+- **More Tracks**: Initialized 'tech_debt_and_test_cleanup_20260302' and 'conductor_workflow_improvements_20260302' to harden TDD discipline, resolve test tech debt (false-positives, dupes), and mandate AST-based codebase auditing.
+- **Final Track**: Initialized 'architecture_boundary_hardening_20260302' to fix the GUI HITL bypass allowing direct AST mutations, patch token bloat in `mma_exec.py`, and implement cascading blockers in `dag_engine.py`.
+- **Testing Consolidation**: Initialized 'testing_consolidation_20260302' track to standardize simulation testing workflows around the pytest `live_gui` fixture and eliminate redundant `subprocess.Popen` wrappers.
+- **Dependency Order**: Added an explicit 'Track Dependency Order' execution guide to `TASKS.md` to ensure safe progression through the accumulated tech debt.
+- **Documentation**: Added guide_meta_boundary.md to explicitly clarify the difference between the Application's strict-HITL environment and the autonomous Meta-Tooling environment, helping future Tiers avoid feature bleed.
+- **Heuristics & Backlog**: Added Data-Oriented Design and Immediate Mode architectural heuristics (inspired by Muratori/Acton) to product-guidelines.md. Logged future decoupling and robust parsing tracks to a 'Future Backlog' in TASKS.md.
+
+---
+
+## 2026-03-02 (Session 3)
+
+### Track: feature_bleed_cleanup_20260302 — Completed |TASK:feature_bleed_cleanup_20260302|
+- **What**: Removed all confirmed dead code and layout regressions from gui_2.py (3 phases)
+- **Why**: Tier 3 workers had left behind dead duplicate methods, dead menu block, duplicate state vars, and a broken Token Budget layout that embedded the panel inside Provider & Model with double labels
+- **How**:
+  - Phase 1: Deleted dead `_render_comms_history_panel` duplicate (stale `type` key, nonexistent `_cb_load_prior_log`, `scroll_area` ID collision). Deleted 4 duplicate `__init__` assignments (ui_new_track_name etc.)
+  - Phase 2: Deleted dead `begin_main_menu_bar()` block (24 lines, always-False in HelloImGui). Added working `Quit` to `_show_menus` via `runner_params.app_shall_exit = True`
+  - Phase 3: Removed 4 redundant Token Budget labels/call from `_render_provider_panel`. Added `collapsing_header("Token Budget")` to AI Settings with proper `_render_token_budget_panel()` call
+- **Issues**: Full test suite hangs (pre-existing — `test_suite_performance_and_flakiness` backlog). Ran targeted GUI/MMA subset (32 passed) as regression proxy. Meta-Level Sanity Check: 52 ruff errors in gui_2.py before and after — zero new violations introduced
+- **Result**: All 3 phases verified by user. Checkpoints: be7174c (Phase 1), 15fd786 (Phase 2), 0d081a2 (Phase 3)
+
+---
+
+## 2026-03-02 (Session 4)
+
+### Track: mma_agent_focus_ux_20260302 — Completed |TASK:mma_agent_focus_ux_20260302|
+- **What**: Per-tier agent focus UX — source_tier tagging + Focus Agent filter UI (all 3 phases)
+- **Why**: All MMA observability panels were global/session-scoped; traffic from Tier 2/3/4 was indistinguishable
+- **How**:
+  - Phase 1: Added `current_tier: str | None` module var to `ai_client.py`; `_append_comms` stamps `source_tier: current_tier` on every comms entry; `run_worker_lifecycle` sets `"Tier 3"` / `generate_tickets` sets `"Tier 2"` around `send()` calls, clears in `finally`; `_on_tool_log` captures `current_tier` at call time; `_append_tool_log` migrated from tuple to dict with `source_tier` field; `_pending_tool_calls` likewise. Checkpoint: bc1a570
+  - Phase 2: `_render_tool_calls_panel` migrated from tuple destructure to dict access. Checkpoint: 865d8dd
+  - Phase 3: `ui_focus_agent: str | None` state var added; Focus Agent combo (All/Tier2/3/4) + clear button above OperationsTabs; filter logic in `_render_comms_history_panel` and `_render_tool_calls_panel`; `[source_tier]` label per comms entry header. Checkpoint: b30e563
+- **Issues**:
+  - `claude_mma_exec.py` fails with nested session block — user authorized inline implementation for this track
+  - Task 2.1 set_file_slice applied at shifted line, leaving stale tuple destructure + missing `i = i_minus_one + 1`; caught and fixed in Phase 3 Task 3.4
+  - **Known limitation**: `current_tier` is a module-level `str | None` — safe only because MMA engine serializes `send()` calls. Concurrent Tier 3/4 agents (future) will require `threading.local()` or per-ticket context passing. Logged to backlog.
+  - **Verification gap noted**: No API hook endpoints expose `ui_focus_agent` state for automated testing. Future tracks should wire widget state to `_settable_fields` for `live_gui` fixture verification. Logged to backlog.
+- **Result**: 18 tests passing. Focus Agent combo visible in Operations Hub. Comms entries show `[main]`/`[Tier N]` labels. Meta-Level Sanity Check: 53 ruff errors in gui_2.py before and after — zero new violations.
+
+---
+
+## 2026-03-02 (Session 5)
+
+### Track: tech_debt_and_test_cleanup_20260302 — Botched / Archived
+- **What**: Attempted to centralize test fixtures and enforce test discipline. 
+- **Issues**: Track was launched with a flawed specification that misidentified critical headless API endpoints as "dead code." While centralized `app_instance` fixtures were successfully deployed, it exposed several zero-assertion tests and exacerbated deep architectural issues with the `asyncio` loop lifecycle, causing widespread `RuntimeError: Event loop is closed` warnings and test hangs.
+- **Result**: Track was aborted and archived. A post-mortem `DEBRIEF.md` was generated.
+
+### Strategic Shift: The Strict Execution Queue
+- **What**: Systematically audited the Future Backlog and converted all pending technical debt into a strict, 9-track, linearly ordered execution queue in `conductor/tracks.md`.
+- **Why**: "Mock-Rot" and stateless Tier 3 entropy. Tier 3 workers were blindly using `unittest.mock.patch` to pass tests without testing integration realities, creating a false sense of security.
+- **How**: 
+  - Defined the "Surgical Spec Protocol" to force Tier 1/2 agents to map exact `WHERE/WHAT/HOW/SAFETY` targets for workers.
+  - Initialized 7 new tracks: `test_stabilization_20260302`, `strict_static_analysis_and_typing_20260302`, `codebase_migration_20260302`, `gui_decoupling_controller_20260302`, `hook_api_ui_state_verification_20260302`, `robust_json_parsing_tech_lead_20260302`, `concurrent_tier_source_tier_20260302`, and `test_suite_performance_and_flakiness_20260302`.
+  - Added a highly interactive `manual_ux_validation_20260302` track specifically for tuning GUI animations and structural layout using a slow-mode simulation harness.
+- **Result**: The project now has a crystal-clear, heavily guarded roadmap to escape technical debt and transition to a robust, Data-Oriented, type-safe architecture.
+## 2026-03-02: Test Suite Stabilization & Simulation Hardening
+*   **Track:** Test Suite Stabilization & Consolidation
+*   **Outcome:** Track Completed Successfully
+*   **Key Accomplishments:**
+    *   **Asyncio Lifecycle Fixes:** Eliminated pervasive Event loop is closed and coroutine was never awaited warnings in tests. Refactored conftest.py teardowns and test loop handling.
+    *   **Legacy Cleanup:** Completely removed gui_legacy.py and updated all 16 referencing test files to target gui_2.py, consolidating the architecture.
+    *   **Functional Assertions:** Replaced pytest.fail placeholders with actual functional assertions in pi_events, execution_engine, 	oken_usage, gent_capabilities, and gent_tools_wiring test suites.
+    *   **Simulation Hardening:** Addressed flakiness in 	est_extended_sims.py. Fixed timeouts and entry count regressions by forcing explicit GUI states (uto_add_history=True) during setup, and refactoring wait_for_ai_response to intelligently detect turn completions and tool execution stalls based on status transitions rather than just counting messages.
+    *   **Workflow Updates:** Updated conductor/workflow.md to establish a new rule forbidding full suite execution (pytest tests/) during verification to prevent long timeouts and threading access violations. Demanded batch-testing (max 4 files) instead.
+    *   **New Track Proposed:** Created sync_tool_execution_20260303 track to introduce concurrent background tool execution, reducing latency during AI research phases.
+*   **Challenges:** The extended simulation suite (	est_extended_sims.py) was highly sensitive to the exact transition timings of the mocked gemini_cli and the background threading of gui_2.py. Required multiple iterations of refinement to simulation/workflow_sim.py to achieve stable, deterministic execution. The full test suite run proved unstable due to accumulation of open threads/loops across 360+ tests, necessitating a shift to batch-testing.

MMA_Support/Architecture_Recommendation.md

@@ -0,0 +1,45 @@
+# MMA Hierarchical Delegation: Recommended Architecture
+
+## 1. Overview
+The Multi-Model Architecture (MMA) utilizes a 4-Tier hierarchy to ensure token efficiency and structural integrity. The primary agent (Conductor) acts as the Tier 2 Tech Lead, delegating specific, stateless tasks to Tier 3 (Workers) and Tier 4 (Utility) agents.
+
+## 2. Agent Roles & Responsibilities
+
+### Tier 2: The Conductor (Tech Lead)
+- **Role:** Orchestrator of the project lifecycle via the Conductor framework.
+- **Context:** High-reasoning, long-term memory of project goals and specifications.
+- **Key Tool:** `mma-orchestrator` skill (Strategy).
+- **Delegation Logic:** Identifies tasks that would bloat the primary context (large code blocks, massive error traces) and spawns sub-agents.
+
+### Tier 3: The Worker (Contributor)
+- **Role:** Stateless code generator.
+- **Context:** Isolated. Sees only the target file and the specific ticket.
+- **Protocol:** Receives a "Worker" system prompt. Outputs clean code or diffs.
+- **Invocation:** `.\scripts\run_subagent.ps1 -Role Worker -Prompt "..."`
+
+### Tier 4: The Utility (QA/Compressor)
+- **Role:** Stateless translator and summarizer.
+- **Context:** Minimal. Sees only the error trace or snippet.
+- **Protocol:** Receives a "QA" system prompt. Outputs compressed findings (max 50 tokens).
+- **Invocation:** `.\scripts\run_subagent.ps1 -Role QA -Prompt "..."`
+
+## 3. Invocation Protocol
+
+### Step 1: Detection
+Tier 2 detects a delegation trigger:
+- Coding task > 50 lines.
+- Error trace > 100 lines.
+
+### Step 2: Spawning
+Tier 2 calls the delegation script:
+```powershell
+.\scripts\run_subagent.ps1 -Role <Worker|QA> -Prompt "Specific instructions..."
+```
+
+### Step 3: Integration
+Tier 2 receives the sub-agent's response.
+- **If Worker:** Tier 2 applies the code changes (using `replace` or `write_file`) and verifies.
+- **If QA:** Tier 2 uses the compressed error to inform the next fix attempt or passes it to a Worker.
+
+## 4. System Prompt Management
+The `run_subagent.ps1` script should be updated to maintain a library of role-specific system prompts, ensuring that Tier 3/4 agents remain focused and tool-free (to prevent nested complexity).

MMA_Support/Data_Pipelines_and_Config.md

@@ -0,0 +1,32 @@
+# Data Pipelines, Memory Views & Configuration
+
+The 4-Tier Architecture relies on strictly managed data pipelines and configuration files to prevent token bloat and maintain a deterministically safe execution environment.
+
+## 1. AST Extraction Pipelines (Memory Views)
+
+To prevent LLMs from hallucinating or consuming massive context windows, raw file text is heavily restricted. The `file_cache.py` uses Tree-sitter for deterministic Abstract Syntax Tree (AST) parsing to generate specific views:
+
+1.  **The Directory Map (Tier 1):** Just filenames and nested paths (e.g., output of `tree /F`). No source code.
+2.  **The Skeleton View (Tier 2 & 3 Dependencies):** Extracts only `class` and `def` signatures, parameters, and type hints. Strips all docstrings and function bodies, replacing them with `pass`. Used for foreign modules a worker must call but not modify.
+3.  **The Curated Implementation View (Tier 2 Target Modules):**
+    *   Keeps class/struct definitions.
+    *   Keeps module-level docstrings and block comments (heuristics).
+    *   Keeps full bodies of functions marked with `@core_logic` or `# [HOT]`.
+    *   Replaces standard function bodies with `... # Hidden`.
+4.  **The Raw View (Tier 3 Target File):** Unredacted, line-by-line source code of the *single* file a Tier 3 worker is assigned to modify.
+
+## 2. Configuration Schema
+
+The architecture separates sensitive billing logic from AI behavior routing.
+
+*   **`credentials.toml` (Security Prerequisite):** Holds the bare metal authentication (`gemini_api_key`, `anthropic_api_key`, `deepseek_api_key`). **This file must be in `.gitignore`.** Loaded strictly for instantiating HTTP clients.
+*   **`project.toml` (Repo Rules):** Holds repository-specific bounds (e.g., "This project uses Python 3.12 and strictly follows PEP8").
+*   **`agents.toml` (AI Routing):** Defines the hardcoded hierarchy's operational behaviors. Includes fallback models (`default_expensive`, `default_cheap`), Tier 1/2 overarching parameters (temperature, base system prompts), and Tier 3 worker archetypes (`refactor`, `codegen`, `contract_stubber`) mapped to specific models (DeepSeek V3, Gemini Flash) and `trust_level` tags (`step` vs. `auto`).
+
+## 3. LLM Output Formats
+
+To ensure robust parser execution and avoid JSON string-escaping nightmares, the architecture uses a hybrid approach for LLM outputs depending on the Tier:
+
+*   **Native Structured Outputs (JSON Schema forced by API):** Used for Tier 1 and Tier 2 routing and orchestration. The model provider mathematically guarantees the syntax, allowing clean parsing of `Track` and `Ticket` metadata by `pydantic`.
+*   **XML Tags (`<file_path>`, `<file_content>`):** Used for Tier 3 Code Generation & Tools. It natively isolates syntax and requires zero string escaping. The UI/Orchestrator parses these via regex to safely extract raw Python code without bracket-matching failures.
+*   **Godot ECS Flat List (Linearized Entities with ID Pointers):** Instead of deeply nested JSON (which models hallucinate across 500 tokens), Tier 1/2 Orchestrators define complex dependency DAGs as a flat list of items (e.g., `[Ticket id="tkt_impl" depends_on="tkt_stub"]`). The Python state machine reconstructs the DAG locally.

MMA_Support/Final_Analysis_Report.md

@@ -0,0 +1,30 @@
+# MMA Tiered Architecture: Final Analysis Report
+
+## 1. Executive Summary
+The implementation and verification of the 4-Tier Hierarchical Multi-Model Architecture (MMA) within the Conductor framework have been successfully completed. The architecture provides a robust "Token Firewall" that prevents the primary context from being bloated by repetitive coding tasks and massive error traces.
+
+## 2. Architectural Findings
+
+### Centralized Strategy vs. Role-Based Sub-Agents
+- **Decision:** A Hybrid Approach was implemented.
+- **Rationale:** The Tier 2 Orchestrator (Conductor) maintains the high-level strategy via a centralized skill, while Tier 3 (Worker) and Tier 4 (QA) agents are governed by surgical, role-specific system prompts. This ensures that sub-agents remain focused and stateless without the overhead of complex, nested tool-usage logic.
+
+### Delegation Efficacy
+- **Tier 3 (Worker):** Successfully isolated code generation from the main conversation. The worker generates clean code/diffs that are then integrated by the Orchestrator.
+- **Tier 4 (QA):** Demonstrated superior token efficiency by compressing multi-hundred-line stack traces into ~20-word actionable fixes.
+- **Traceability:** The `-ShowContext` flag in `scripts/run_subagent.ps1` provides immediate visibility into the "Connective Tissue" of the hierarchy, allowing human supervisors to monitor the hand-offs.
+
+## 3. Recommended Protocol (Final)
+
+1. **Identification:** Tier 2 identifies a "Bloat Trigger" (Coding > 50 lines, Errors > 100 lines).
+2. **Delegation:** Tier 2 spawns a sub-agent via `.\scripts
+un_subagent.ps1 -Role [Worker|QA] -Prompt "..."`.
+3. **Integration:** Tier 2 receives the stateless response and applies it to the project state.
+4. **Checkpointing:** Tier 2 performs Phase-level checkpoints to "Wipe" trial-and-error memory and solidify the new state.
+
+## 4. Verification Results
+- **Automated Tests:** 100% Pass (4/4 tests in `tests/conductor/test_infrastructure.py`).
+- **Isolation:** Confirmed via `test_subagent_isolation_live`.
+- **Live Trace:** Manually verified and approved by the user (Tier 2 -> 3 -> 4 flow).
+
+## 5. Conclusion

MMA_Support/Implementation_Tracks.md

@@ -0,0 +1,46 @@
+# Iteration Plan (Implementation Tracks)
+
+To safely refactor a linear, single-agent codebase into the 4-Tier Multi-Model Architecture without breaking the working prototype, the implementation should be sequenced into these five isolated Epics (Tracks):
+
+## Track 1: The Memory Foundations (AST Parser)
+**Goal:** Build the engine that prevents token-bloat by turning massive source files into curated memory views.
+**Implementation Details:**
+1.  Integrate `tree-sitter` and language bindings into `file_cache.py`.
+2.  Build `ASTParser` extraction rules:
+    *   *Skeleton View:* Strip function/class bodies, preserving only signatures, parameters, and type hints.
+    *   *Curated View:* Preserve class structures, module docstrings, and bodies of functions marked `# [HOT]` or `@core_logic`. Replace standard bodies with `... # Hidden`.
+3.  **Acceptance:** `file_cache.get_curated_view('script.py')` returns a perfectly formatted summary string in the terminal.
+
+## Track 2: State Machine & Data Structures
+**Goal:** Define the rigid Python objects the AI agents will pass to each other to rely on structured data, not loose chat strings.
+**Implementation Details:**
+1.  Create `models.py` with `pydantic` or `dataclasses` for `Track` (Epic) and `Ticket` (Task).
+2.  Define `WorkerContext` holding the Ticket ID, assigned model (from `agents.toml`), isolated `credentials.toml` injection, and a `messages` payload array.
+3.  Add helper methods for state mutators (e.g., `ticket.mark_blocked()`, `ticket.mark_complete()`).
+4.  **Acceptance:** Instantiate a `Track` with 3 `Tickets` and successfully enforce state changes in Python without AI involvement.
+
+## Track 3: The Linear Orchestrator & Execution Clutch
+**Goal:** Build the synchronous, debuggable core loop that runs a single Tier 3 Worker and pauses for human approval.
+**Implementation Details:**
+1.  Create `multi_agent_conductor.py` with a `run_worker_lifecycle(ticket: Ticket)` function.
+2.  Inject context (Raw View from `file_cache.py`) and format the `messages` array for the API.
+3.  Implement the Clutch (HITL): `input()` pause for CLI or wait state for GUI before executing the returned tool (e.g., `write_file`). Allow manual memory mutation of the JSON payload.
+4.  **Acceptance:** The script sends a hardcoded Ticket to DeepSeek, pauses in the terminal showing a diff, waits for user approval, applies the diff via `mcp_client.py`, and wipes the worker's history.
+
+## Track 4: Tier 4 QA Interception
+**Goal:** Stop error traces from destroying the Worker's token window by routing crashes through a stateless translator.
+**Implementation Details:**
+1.  In `shell_runner.py`, intercept `stderr` (e.g., `returncode != 0`).
+2.  Do *not* append `stderr` to the main Worker's history. Instead, instantiate a synchronous API call to the `default_cheap` model.
+3.  Prompt: *"You are an error parser. Output only a 1-2 sentence instruction on how to fix this syntax error."* Send the raw `stderr` and target file snippet.
+4.  Append the translated 20-word fix to the main Worker's history as a "System Hint".
+5.  **Acceptance:** A deliberate syntax error triggers the execution engine to silently ping the cheap API, returning a 20-word correction to the Worker instead of a 200-line stack trace.
+
+## Track 5: UI Decoupling & Tier 1/2 Routing (The Final Boss)
+**Goal:** Bring the system online by letting Tier 1 and Tier 2 dynamically generate Tickets managed by the async Event Bus.
+**Implementation Details:**
+1.  Implement an `asyncio.Queue` in `multi_agent_conductor.py`.
+2.  Write Tier 1 & 2 system prompts forcing output as strict JSON arrays (Tracks and Tickets).
+3.  Write the Dispatcher async loop to convert JSON into `Ticket` objects and push to the queue.
+4.  Enforce the Stub Resolver: If a Ticket archetype is `contract_stubber`, pause dependent Tickets, run the stubber, trigger `file_cache.py` to rebuild the Skeleton View, then resume.
+5.  **Acceptance:** Vague prompt ("Refactor config system") results in Tier 1 Track, Tier 2 Tickets (Interface stub + Implementation). System executes stub, updates AST, and finishes implementation automatically (or steps through if Linear toggle is on).

MMA_Support/Orchestrator_Engine.md

@@ -0,0 +1,37 @@
+# The Orchestrator Engine & UI
+
+To transition from a linear, single-agent chat box to a multi-agent control center, the GUI must be decoupled from the LLM execution loops. A single-agent UI assumes a linear flow (*User types -> UI waits -> LLM responds -> UI updates*), which freezes the application if a Tier 1 PM waits for human approval while Tier 3 Workers run local tests in the background.
+
+## 1. The Async Event Bus (Decoupling UI from Agents)
+
+The GUI acts as a "dumb" renderer. It only renders state; it never manages state.
+
+*   **The Agent Bus (Message Queue):** A thread-safe signaling system (e.g., `asyncio.Queue`, `pyqtSignal`) passes messages between agents, UI, and the filesystem.
+*   **Background Workers:** When Tier 1 spawns a Tier 2 Tech Lead, the GUI does not wait. It pushes a `UserRequestEvent` to the Conductor's queue. The Conductor runs the LLM call asynchronously and fires `StateUpdateEvents` back for the GUI to redraw.
+
+## 2. The Execution Clutch (HITL)
+
+Every spawned worker panel implements an execution state toggle based on the `trust_level` defined in `agents.toml`.
+
+*   **Step Mode (Lock-step):** The worker pauses **twice** per cycle:
+    1.  *After* generating a response/tool-call, but *before* executing the tool. The GUI renders a preview (e.g., diff of lines 40-50) and offers `[Approve]`, `[Edit Payload]`, or `[Abort]`.
+    2.  *After* executing the tool, but *before* sending output back to the LLM (allows verification of the system output).
+*   **Auto Mode (Fire-and-forget):** The worker loops continuously until it outputs a "Task Complete" status to the Router.
+
+## 3. Memory Mutation (The "Debug" Superpower)
+
+If a worker generates a flawed plan in Step Mode, the "Memory Mutator" allows the user to click the last message and edit the raw JSON/text directly before hitting "Approve." By rewriting the AI's brain mid-task, the model proceeds as if it generated the correct idea, saving the context window from restarting due to a minor hallucination.
+
+## 4. The Global Execution Toggle
+
+A Global Execution Toggle overrides all individual agent trust levels for debugging race conditions or context leaks.
+
+*   **Mode = "async" (Production):** The Dispatcher throws Tickets into an `asyncio.TaskGroup`. They spawn instantly, fight for API rate limits, read the skeleton, and run in parallel.
+*   **Mode = "linear" (Debug):** The Dispatcher iterates through the array sequentially using a strict `for` loop. It `awaits` absolute completion of Ticket 1 (including QA loops and code review) before instantiating the `WorkerAgent` for Ticket 2. This enforces a deterministic state machine and outputs state snapshots (`debug_state.json`) for manual verification.
+
+## 5. State Machine (Dataclasses)
+
+The Conductor relies on strict definitions for `Track` and `Ticket` to enforce state and UI rendering (e.g., using `dataclasses` or `pydantic`).
+
+*   **`Ticket`:** Contains `id`, `target_file`, `prompt`, `worker_archetype`, `status` (pending, running, blocked, step_paused, completed), and a `dependencies` list of Ticket IDs that must finish first.
+*   **`Track`:** Contains `id`, `title`, `description`, `status`, and a list of `Tickets`.

MMA_Support/Overview.md

@@ -0,0 +1,18 @@
+# System Specification: 4-Tier Hierarchical Multi-Model Architecture
+
+**Project:** `manual_slop` (or equivalent Agentic Co-Dev Prototype)
+
+**Core Philosophy:** Token Economy, Strict Memory Siloing, and Human-In-The-Loop (HITL) Execution.
+
+## 1. Architectural Overview
+
+This system rejects the "monolithic black-box" approach to agentic coding. Instead of passing an entire codebase into a single expensive context window, the architecture mimics a senior engineering department. It uses a 4-Tier hierarchy where cognitive load and context are aggressively filtered from top to bottom.
+
+Expensive, high-reasoning models manage metadata and architecture (Tier 1 & 2), while cheap, fast models handle repetitive syntax and error parsing (Tier 3 & 4).
+
+### 1.1 Core Paradigms
+
+* **Token Firewalling:** Error logs and deep history are never allowed to bubble up to high-tier models. The system relies heavily on abstracted AST views (Skeleton, Curated) rather than raw code when context allows.
+* **Context Amnesia:** Worker agents (Tier 3) have their trial-and-error histories wiped upon task completion to prevent context ballooning and hallucination.
+* **The Execution Clutch (HITL):** Agents operate based on Archetype Trust Scores defined in configuration. Trusted patterns run in `Auto` mode; untrusted or complex refactors run in `Step` mode, pausing before tool execution for human review and JSON history mutation.
+* **Interface-Driven Development (IDD):** The architecture inherently prioritizes the creation of contracts (stubs, schemas) before implementation, allowing workers to proceed in parallel without breaking cross-module boundaries.

MMA_Support/Tier1_Orchestrator.md

@@ -0,0 +1,38 @@
+# Tier 1: The Top-Level Orchestrator (Product Manager)
+
+**Designated Models:** Gemini 3.1 Pro, Claude 3.5 Sonnet.
+**Execution Frequency:** Low (Start of feature, Macro-merge resolution).
+**Core Role:** Epic planning, architecture enforcement, and cross-module task delegation.
+
+The Tier 1 Orchestrator is the most capable and expensive model in the hierarchy. It operates strictly on metadata, summaries, and executive-level directives. It **never** sees raw implementation code.
+
+## Memory Context & Paths
+
+### Path A: Epic Initialization (Project Planning)
+*   **Trigger:** User drops a massive new feature request or architectural shift into the main UI.
+*   **What it Sees (Context):**
+    *   **The User Prompt:** The raw feature request.
+    *   **Project Meta-State:** `project.toml` (rules, allowed languages, dependencies).
+    *   **Repository Map:** A strict, file-tree outline (names and paths only).
+    *   **Global Architecture Docs:** High-level markdown files (e.g., `docs/guide_architecture.md`).
+*   **What it Ignores:** All source code, all AST skeletons, and all previous micro-task histories.
+*   **Output Format:** A JSON array (Godot ECS Flat List format) of `Tracks` (Jira Epics), identifying which modules will be affected, the required Tech Lead persona, and the severity level.
+
+### Path B: Track Delegation (Sprint Kickoff)
+*   **Trigger:** The PM is handing a defined Track down to a Tier 2 Tech Lead.
+*   **What it Sees (Context):**
+    *   **The Target Track:** The specific goal and Acceptance Criteria generated in Path A.
+    *   **Module Interfaces (Skeleton View):** Strict AST skeleton (just class/function definitions) *only* for the modules this specific Track is allowed to touch.
+    *   **Track Roster:** A list of currently active or completed Tracks to prevent duplicate work.
+*   **What it Ignores:** Unrelated module docs, original massive user prompt, implementation details.
+*   **Output Format:** A compiled "Track Brief" (system prompt + curated file list) passed to instantiate the Tier 2 Tech Lead panel.
+
+### Path C: Macro-Merge & Acceptance Review (Severity Resolution)
+*   **Trigger:** A Tier 2 Tech Lead reports "Track Complete" and submits a pull request/diff for a "High Severity" task.
+*   **What it Sees (Context):**
+    *   **Original Acceptance Criteria:** The Track's goals.
+    *   **Tech Lead's Executive Summary:** A ~200-word explanation of the chosen implementation algorithm.
+    *   **The Macro-Diff:** Actual changes made to the codebase.
+    *   **Curated Implementation View:** For boundary files, ensuring the merge doesn't break foreign modules.
+*   **What it Ignores:** Tier 3 Worker trial-and-error histories, Tier 4 error logs, raw bodies of unchanged functions.
+*   **Output Format:** "Approved" (commits to memory) OR "Rejected" with specific architectural feedback for Tier 2.

MMA_Support/Tier2_TechLead.md

@@ -0,0 +1,46 @@
+# Tier 2: The Track Conductor (Tech Lead)
+
+**Designated Models:** Gemini 3.0 Flash, Gemini 2.5 Pro.
+**Execution Frequency:** Medium.
+**Core Role:** Module-specific planning, code review, spawning Worker agents, and Topological Dependency Graph management.
+
+The Tech Lead bridges the gap between high-level architecture and actual code syntax. It operates in a "need-to-know" state, utilizing AST parsing (`file_cache.py`) to keep token counts low while maintaining structural awareness of its assigned modules.
+
+## Memory Context & Paths
+
+### Path A: Sprint Planning (Task Delegation)
+*   **Trigger:** Tier 1 (PM) assigns a Track (Epic) and wakes up the Tech Lead.
+*   **What it Sees (Context):**
+    *   **The Track Brief:** Acceptance Criteria from Tier 1.
+    *   **Curated Implementation View (Target Modules):** AST-extracted class structures, docstrings, and `# [HOT]` function bodies for the 1-3 files this Track explicitly modifies.
+    *   **Skeleton View (Foreign Modules):** Only function signatures and return types for external dependencies.
+*   **What it Ignores:** The rest of the repository, the PM's overarching project-planning logic, raw line-by-line code of non-hot functions.
+*   **Output Format:** A JSON array (Godot ECS Flat List format) of discrete Tier 3 `Tickets` (e.g., Ticket 1: *Write DB migration script*, Ticket 2: *Update core API endpoints*), including `depends_on` pointers to construct an execution DAG.
+
+### Path B: Code Review (Local Integration)
+*   **Trigger:** A Tier 3 Contributor completes a Ticket and submits a diff, OR Tier 4 (QA) flags a persistent failure.
+*   **What it Sees (Context):**
+    *   **Specific Ticket Goal:** What the Contributor was instructed to do.
+    *   **Proposed Diff:** The exact line changes submitted by Tier 3.
+    *   **Test/QA Output:** Relevant logs from Tier 4 compiler checks.
+    *   **Curated Implementation View:** To cross-reference the proposed diff against the existing architecture.
+*   **What it Ignores:** The Contributor's internal trial-and-error chat history. It only sees the final submission.
+*   **Output Format:** *Approve* (merges diff into working branch and updates Curated View) or *Reject* (sends technical critique back to Tier 3).
+
+### Path C: Track Finalization (Upward Reporting)
+*   **Trigger:** All Tier 3 Tickets assigned to this Track are marked "Approved."
+*   **What it Sees (Context):**
+    *   **Original Track Brief:** To verify requirements were met.
+    *   **Aggregated Track Diff:** The sum total of all changes made across all Tier 3 Tickets.
+    *   **Dependency Delta:** A list of any new foreign modules or libraries imported.
+*   **What it Ignores:** The back-and-forth review cycles, original AST Curated View.
+*   **Output Format:** An Executive Summary and the final Macro-Diff, sent back to Tier 1.
+
+### Path D: Contract-First Delegation (Stub-and-Resolve)
+*   **Trigger:** Tier 2 evaluates a Track and detects a cross-module dependency (or a single massive refactor) requiring an undefined signature.
+*   **Role:** Force Interface-Driven Development (IDD) to prevent hallucination.
+*   **Execution Flow:**
+    1.  **Contract Definition:** Splits requirement into a `Stub Ticket`, `Consumer Ticket`, and `Implementation Ticket`.
+    2.  **Stub Generation:** Spawns a cheap Tier 3 worker (e.g., DeepSeek V3 `contract_stubber` archetype) to generate the empty function signature, type hints, and docstrings.
+    3.  **Skeleton Broadcast:** The stub merges, and the system instantly re-runs Tree-sitter to update the global Skeleton View.
+    4.  **Parallel Implementation:** Tier 2 simultaneously spawns the `Consumer` (codes against the skeleton) and the `Implementer` (fills the stub logic) in isolated contexts.

MMA_Support/Tier3_Workers.md

@@ -0,0 +1,35 @@
+# Tier 3: The Worker Agents (Contributors)
+
+**Designated Models:** DeepSeek V3/R1, Gemini 2.5 Flash.
+**Execution Frequency:** High (The core loop).
+**Core Role:** Generating syntax, writing localized files, running unit tests.
+
+The engine room of the system. Contributors execute the highest volume of API calls. Their memory context is ruthlessly pruned. By leveraging cheap, fast models, they operate with zero architectural anxiety—they just write the code they are assigned. They are "Amnesiac Workers," having their history wiped between tasks to prevent context ballooning.
+
+## Memory Context & Paths
+
+### Path A: Heads Down Execution (Task Execution)
+*   **Trigger:** Tier 2 (Tech Lead) hands down a hyper-specific Ticket.
+*   **What it Sees (Context):**
+    *   **The Ticket Prompt:** The exact, isolated instructions from Tier 2.
+    *   **The Target File (Raw View):** The raw, unredacted, line-by-line source code of *only* the specific file (or class/function) it was assigned to modify.
+    *   **Foreign Interfaces (Skeleton View):** Strict AST skeleton (signatures only) of external dependencies required by the ticket.
+*   **What it Ignores:** Epic/Track goals, Tech Lead's Curated View, other files in the same directory, parallel Tickets.
+*   **Output Format:** XML Tags (`<file_path>`, `<file_content>`) defining direct file modifications or `mcp_client.py` tool payloads.
+
+### Path B: Trial and Error (Local Iteration & Tool Execution)
+*   **Trigger:** The Contributor runs a local linter/test, encounters a syntax error, or the human pauses execution using "Step" mode.
+*   **What it Sees (Context):**
+    *   **Ephemeral Working History:** A short, rolling window of its last 2–3 attempts (e.g., "Attempt 1: Wrote code -> Tool Output: SyntaxError").
+    *   **Tier 4 (QA) Injections:** Compressed (20-50 token) fix recommendations from Tier 4 agents (e.g., "Add a closing bracket on line 42").
+    *   **Human Mutations:** Any direct edits made to its JSON history payload before proceeding.
+*   **What it Ignores:** Tech Lead code reviews, attempts older than the rolling window (wiped to save tokens).
+*   **Output Format:** Revised tool payloads until tests pass or the human approves.
+
+### Path C: Task Submission (Micro-Pull Request)
+*   **Trigger:** The code executes cleanly, and "Step" mode is finalized into "Task Complete."
+*   **What it Sees (Context):**
+    *   **The Original Ticket:** To confirm instructions were met.
+    *   **The Final State:** The cleanly modified file or exact diff.
+*   **What it Ignores:** **All of Path B.** Before submission to Tier 2, the orchestrator wipes the messy trial-and-error history from the payload.
+*   **Output Format:** A concise completion message and the clean diff, sent up to Tier 2.

MMA_Support/Tier4_Utility.md

@@ -0,0 +1,33 @@
+# Tier 4: The Utility Agents (Compiler / QA)
+
+**Designated Models:** DeepSeek V3 (Lowest cost possible).
+**Execution Frequency:** On-demand (Intercepts local failures).
+**Core Role:** Single-shot, stateless translation of machine garbage into human English.
+
+Tier 4 acts as the financial firewall. It solves the expensive problem of feeding massive (e.g., 3,000-token) stack traces back into a mid-tier LLM's context window. Tier 4 agents wake up, translate errors, and immediately die.
+
+## Memory Context & Paths
+
+### Path A: The Stack Trace Interceptor (Translator)
+*   **Trigger:** A Tier 3 Contributor executes a script, resulting in a non-zero exit code with a massive `stderr` payload.
+*   **What it Sees (Context):**
+    *   **Raw Error Output:** The exact traceback from the runtime/compiler.
+    *   **Offending Snippet:** *Only* the specific function or 20-line block of code where the error originated.
+*   **What it Ignores:** Everything else. It is blind to the "Why" and focuses only on "What broke."
+*   **Output Format:** A surgical, highly compressed string (20-50 tokens) passed back into the Tier 3 Contributor's working memory (e.g., "Syntax Error on line 42: You missed a closing parenthesis. Add `]`").
+
+### Path B: The Linter / Formatter (Pedant)
+*   **Trigger:** Tier 3 believes it finished a Ticket, but pre-commit hooks (e.g., `ruff`, `eslint`) fail.
+*   **What it Sees (Context):**
+    *   **Linter Warning:** Specific error (e.g., "Line too long", "Missing type hint").
+    *   **Target File:** Code written by Tier 3.
+*   **What it Ignores:** Business logic. It only cares about styling rules.
+*   **Output Format:** A direct `sed` command or silent diff overwrite via tools to fix the formatting without bothering Tier 2 or consuming Tier 3 loops.
+
+### Path C: The Flaky Test Debugger (Isolator)
+*   **Trigger:** A localized unit test fails due to logic (e.g., `assert 5 == 4`), not a syntax crash.
+*   **What it Sees (Context):**
+    *   **Failing Test Function:** The exact `pytest` or `go test` block.
+    *   **Target Function:** The specific function being tested.
+*   **What it Ignores:** The rest of the test suite and module.
+*   **Output Format:** A quick diagnosis sent to Tier 3 (e.g., "The test expects an integer, but your function is currently returning a stringified float. Cast to `int`").

MMA_Support/mma_tiered_orchestrator_skill.md

@@ -0,0 +1,66 @@
+# Skill: MMA Tiered Orchestrator
+
+## Description
+This skill enforces the 4-Tier Hierarchical Multi-Model Architecture (MMA) directly within the Gemini CLI using Token Firewalling and sub-agent task delegation. It teaches the CLI how to act as a Tier 1/2 Orchestrator, dispatching stateless tasks to cheaper models using shell commands, thereby preventing massive error traces or heavy coding contexts from polluting the primary prompt context.
+
+<instructions>
+# MMA Token Firewall & Tiered Delegation Protocol
+
+You are operating as a Tier 1 Product Manager or Tier 2 Tech Lead within the MMA Framework. Your context window is extremely valuable and must be protected from token bloat (such as raw, repetitive code edits, trial-and-error histories, or massive stack traces).
+
+To accomplish this, you MUST delegate token-heavy or stateless tasks to "Tier 3 Contributors" or "Tier 4 QA Agents" by spawning secondary Gemini CLI instances via `run_shell_command`.
+
+**CRITICAL Prerequisite:**
+To avoid hanging the CLI and ensure proper environment authentication, you MUST NOT call the `gemini` command directly. Instead, you MUST use the wrapper script:
+`.\scripts\run_subagent.ps1 -Prompt "..."`
+
+## 1. The Tier 3 Worker (Heads-Down Coding)
+When you need to perform a significant code modification (e.g., refactoring a 500-line script, writing a massive class, or implementing a predefined spec):
+1. **DO NOT** attempt to write or use `replace`/`write_file` yourself. Your history will bloat.
+2. **DO** construct a single, highly specific prompt.
+3. **DO** spawn a sub-agent using `run_shell_command` pointing to the target file.
+   *Command:* `.\scripts\run_subagent.ps1 -Prompt "Modify [FILE_PATH] to implement [SPECIFIC_INSTRUCTION]. Only write the code, no pleasantries."`
+4. If you need the sub-agent to automatically apply changes instead of just returning the text, use `gemini run` or pipe the output appropriately. However, the best method is to let the sub-agent modify the code and return "Done."
+
+## 2. The Tier 4 QA Agent (Error Translation)
+If you run a local test (e.g., `npm test`, `pytest`, `go run`) via `run_shell_command` and it fails with a massive traceback (e.g., 200+ lines of `stderr`):
+1. **DO NOT** analyze the raw `stderr` in your own context window.
+2. **DO** immediately spawn a stateless Tier 4 agent to compress the error.
+3. *Command:* `.\scripts\run_subagent.ps1 -Prompt "Summarize this stack trace into a 20-word fix: [PASTE_SNIPPET_OF_STDERR_HERE]"`
+4. Use the 20-word fix returned by the Tier 4 agent to inform your next architectural decision or pass it to the Tier 3 worker.
+
+## 3. Context Amnesia (Phase Checkpoints)
+When you complete a major Phase or Track within the `conductor` workflow:
+1. Stage your changes and commit them.
+2. Draft a comprehensive summary of the state changes in a Git Note attached to the commit.
+3. Treat the checkpoint as a "Memory Wipe." Actively disregard previous conversational turns and trial-and-error histories. Rely exclusively on the newly generated Git Note and the physical state of the files on disk for your next Phase.
+</instructions>
+
+<examples>
+### Example 1: Spawning a Tier 4 QA Agent
+**User / System:** `pytest tests/test_gui.py` failed with 400 lines of output.
+**Agent (You):** 
+```json
+{
+  "command": ".\\scripts\\run_subagent.ps1 -Prompt \"Summarize this stack trace into a 20-word fix: [snip first 30 lines...]\"",
+  "description": "Spawning Tier 4 QA to compress error trace statelessly."
+}
+```
+
+### Example 2: Spawning a Tier 3 Worker
+**User:** Please implement the `ASTParser` class in `file_cache.py` as defined in Track 1.
+**Agent (You):**
+```json
+{
+  "command": ".\\scripts\\run_subagent.ps1 -Prompt \"Read file_cache.py and implement the ASTParser class using tree-sitter. Ensure you preserve docstrings but strip function bodies. Output the updated code or edit the file directly.\"",
+  "description": "Delegating implementation to a Tier 3 Worker."
+}
+```
+</examples>
+
+<triggers>
+- When asked to write large amounts of boilerplate or repetitive code.
+- When encountering a large error trace from a shell execution.
+- When explicitly instructed to act as a "Tech Lead" or "Orchestrator".
+- When managing complex, multi-file Track implementations.
+</triggers>

MMA_UX_SPEC.md

@@ -0,0 +1,36 @@
+# MMA Observability & UX Specification
+
+## 1. Goal
+Implement the visible surface area of the 4-Tier Hierarchical Multi-Model Architecture within `gui_2.py`. This ensures the user can monitor, control, and debug the multi-agent execution flow.
+
+## 2. Core Components
+
+### 2.1 MMA Dashboard Panel
+- **Visibility:** A new dockable panel named "MMA Dashboard".
+- **Track Status:** Display the current active `Track` ID and overall progress (e.g., "3/10 Tickets Complete").
+- **Ticket DAG Visualization:** A list or simple graph representing the `Ticket` queue.
+    - Each ticket shows: `ID`, `Target`, `Status` (Pending, Running, Paused, Complete, Blocked).
+    - Visual indicators for dependencies (e.g., indented or linked).
+
+### 2.2 The Execution Clutch (HITL)
+- **Step Mode Toggle:** A global or per-track checkbox to enable "Step Mode".
+- **Pause Points:**
+    - **Pre-Execution:** When a Tier 3 worker generates a tool call (e.g., `write_file`), the engine pauses.
+    - **UI Interaction:** The GUI displays the proposed script/change and provides:
+        - `[Approve]`: Proceed with execution.
+        - `[Edit Payload]`: Open the Memory Mutator.
+        - `[Abort]`: Mark the ticket as Blocked/Cancelled.
+- **Visual Feedback:** Tactile/Arcade-style blinking or color changes when the engine is "Paused for HITL".
+
+### 2.3 Memory Mutator (The "Debug" Superpower)
+- **Functionality:** A modal or dedicated text area that allows the user to edit the raw JSON conversation history of a paused worker.
+- **Use Case:** Fixing AI hallucinations or providing specific guidance mid-turn without restarting the context window.
+- **Integration:** After editing, the "Approve" button sends the *modified* history back to the engine.
+
+### 2.4 Tiered Metrics & Logs
+- **Observability:** Show which model (Tier 1, 2, 3, or 4) is currently active.
+- **Sub-Agent Logs:** Provide quick links to open the timestamped log files generated by `mma_exec.py`.
+
+## 3. Technical Integration
+- **Event Bus:** Use the existing `AsyncEventQueue` to push `StateUpdateEvents` from the `ConductorEngine` to the GUI.
+- **Non-Blocking:** Ensure the UI remains responsive (FPS > 60) even when multiple tickets are processing or the engine is waiting for user input.

MainContext.md

@@ -1,266 +0,0 @@
-# Manual Slop
-
-## Summary
-
-Is a local GUI tool for manually curating and sending context to AI APIs. It aggregates files, screenshots, and discussion history into a structured markdown file and sends it to a chosen AI provider with a user-written message. The AI can also execute PowerShell scripts within the project directory, with user confirmation required before each execution.
-
-**Stack:**
-- `dearpygui` - GUI with docking/floating/resizable panels
-- `google-genai` - Gemini API
-- `anthropic` - Anthropic API
-- `tomli-w` - TOML writing
-- `uv` - package/env management
-
-**Files:**
-- `gui.py` - main GUI, `App` class, all panels, all callbacks, confirmation dialog, layout persistence, rich comms rendering
-- `ai_client.py` - unified provider wrapper, model listing, session management, send, tool/function-call loop, comms log, provider error classification
-- `aggregate.py` - reads config, collects files/screenshots/discussion, writes numbered `.md` files to `output_dir`
-- `shell_runner.py` - subprocess wrapper that runs PowerShell scripts sandboxed to `base_dir`, returns stdout/stderr/exit code as a string
-- `session_logger.py` - opens timestamped log files at session start; writes comms entries as JSON-L and tool calls as markdown; saves each AI-generated script as a `.ps1` file
-- `project_manager.py` - per-project .toml load/save, entry serialisation (entry_to_str/str_to_entry with @timestamp support), default_project/default_discussion factories, migrate_from_legacy_config, flat_config for aggregate.run(), git helpers (get_git_commit, get_git_log)
-- `theme.py` - palette definitions, font loading, scale, load_from_config/save_to_config
-- `gemini.py` - legacy standalone Gemini wrapper (not used by the main GUI; superseded by `ai_client.py`)
-- `file_cache.py` - stub; Anthropic Files API path removed; kept so stale imports don't break
-- `mcp_client.py` - MCP-style read-only file tools (read_file, list_directory, search_files, get_file_summary); allowlist enforced against project file_items + base_dirs; dispatched by ai_client tool-use loop for both Anthropic and Gemini
-- `summarize.py` - local heuristic summariser (no AI); .py via AST, .toml via regex, .md headings, generic preview; used by mcp_client.get_file_summary and aggregate.build_summary_section
-- `config.toml` - global-only settings: [ai] provider+model+system_prompt, [theme] palette+font+scale, [projects] paths array + active path
-- `manual_slop.toml` - per-project file: [project] name+git_dir+system_prompt+main_context, [output] namespace+output_dir, [files] base_dir+paths, [screenshots] base_dir+paths, [discussion] roles+active+[discussion.discussions.<name>] git_commit+last_updated+history
-- `credentials.toml` - gemini api_key, anthropic api_key
-- `dpg_layout.ini` - Dear PyGui window layout file (auto-saved on exit, auto-loaded on startup); gitignore this per-user
-
-**GUI Panels:**
-- **Projects** - active project name display (green), git directory input + Browse button, scrollable list of loaded project paths (click name to switch, x to remove), Add Project / New Project / Save All buttons
-- **Config** - namespace, output dir, save (these are project-level fields from the active .toml)
-- **Files** - base_dir, scrollable path list with remove, add file(s), add wildcard
-- **Screenshots** - base_dir, scrollable path list with remove, add screenshot(s)
-- **Discussion History** - discussion selector (collapsible header): listbox of named discussions, git commit + last_updated display, Update Commit button, Create/Rename/Delete buttons with name input; structured entry editor: each entry has collapse toggle (-/+), role combo, timestamp display, multiline content field; per-entry Ins/Del buttons when collapsed; global toolbar: + Entry, -All, +All, Clear All, Save; collapsible **Roles** sub-section; -> History buttons on Message and Response panels append current message/response as new entry with timestamp
-- **Provider** - provider combo (gemini/anthropic), model listbox populated from API, fetch models button
-- **Message** - multiline input, Gen+Send button, MD Only button, Reset session button, -> History button
-- **Response** - readonly multiline displaying last AI response, -> History button
-- **Tool Calls** - scrollable log of every PowerShell tool call the AI made; Clear button
-- **System Prompts** - global (all projects) and project-specific multiline text areas for injecting custom system instructions. Combined with the built-in tool prompt.
-- **Comms History** - rich structured live log of every API interaction; status line at top; colour legend; Clear button
-
-**Layout persistence:**
-- `dpg.configure_app(..., init_file="dpg_layout.ini")` loads the ini at startup if it exists; DPG silently ignores a missing file
-- `dpg.save_init_file("dpg_layout.ini")` is called immediately before `dpg.destroy_context()` on clean exit
-- The ini records window positions, sizes, and dock node assignments in DPG's native format
-- First run (no ini) uses the hardcoded `pos=` defaults in `_build_ui()`; after that the ini takes over
-- Delete `dpg_layout.ini` to reset to defaults
-
-**Project management:**
-- `config.toml` is global-only: `[ai]`, `[theme]`, `[projects]` (paths list + active path). No project data lives here.
-- Each project has its own `.toml` file (e.g. `manual_slop.toml`). Multiple project tomls can be registered by path.
-- `App.__init__` loads global config, then loads the active project `.toml` via `project_manager.load_project()`. Falls back to `migrate_from_legacy_config()` if no valid project file exists, creating a new `.toml` automatically.
-- `_flush_to_project()` pulls widget values into `self.project` (the per-project dict) and serialises disc_entries into the active discussion's history list
-- `_flush_to_config()` writes global settings ([ai], [theme], [projects]) into `self.config`
-- `_save_active_project()` writes `self.project` to the active `.toml` path via `project_manager.save_project()`
-- `_do_generate()` calls both flush methods, saves both files, then uses `project_manager.flat_config()` to produce the dict that `aggregate.run()` expects — so `aggregate.py` needs zero changes
-- Switching projects: saves current project, loads new one, refreshes all GUI state, resets AI session
-- New project: file dialog for save path, creates default project structure, saves it, switches to it
-
-**Discussion management (per-project):**
-- Each project `.toml` stores one or more named discussions under `[discussion.discussions.<name>]`
-- Each discussion has: `git_commit` (str), `last_updated` (ISO timestamp), `history` (list of serialised entry strings)
-- `active` key in `[discussion]` tracks which discussion is currently selected
-- Creating a discussion: adds a new empty discussion dict via `default_discussion()`, switches to it
-- Renaming: moves the dict to a new key, updates `active` if it was the current one
-- Deleting: removes the dict; cannot delete the last discussion; switches to first remaining if active was deleted
-- Switching: flushes current entries to project, loads new discussion's history, rebuilds disc list
-- Update Commit button: runs `git rev-parse HEAD` in the project's `git_dir` and stores result + timestamp in the active discussion
-- Timestamps: each disc entry carries a `ts` field (ISO datetime); shown next to the role combo; new entries from `-> History` or `+ Entry` get `now_ts()`
-
-**Entry serialisation (project_manager):**
-- `entry_to_str(entry)` → `"@<ts>\n<role>:\n<content>"` (or `"<role>:\n<content>"` if no ts)
-- `str_to_entry(raw, roles)` → parses optional `@<ts>` prefix, then role line, then content; returns `{role, content, collapsed, ts}`
-- Round-trips correctly through TOML string arrays; handles legacy entries without timestamps
-
-**AI Tool Use (PowerShell):**
-- Both Gemini and Anthropic are configured with a `run_powershell` tool/function declaration
-- When the AI wants to edit or create files it emits a tool call with a `script` string
-- `ai_client` runs a loop (max `MAX_TOOL_ROUNDS = 10`) feeding tool results back until the AI stops calling tools
-- Before any script runs, `gui.py` shows a modal `ConfirmDialog` on the main thread; the background send thread blocks on a `threading.Event` until the user clicks Approve or Reject
-- The dialog displays `base_dir`, shows the script in an editable text box (allowing last-second tweaks), and has Approve & Run / Reject buttons
-- On approval the (possibly edited) script is passed to `shell_runner.run_powershell()` which prepends `Set-Location -LiteralPath '<base_dir>'` and runs it via `powershell -NoProfile -NonInteractive -Command`
-- stdout, stderr, and exit code are returned to the AI as the tool result
-- Rejections return `"USER REJECTED: command was not executed"` to the AI
-- All tool calls (script + result/rejection) are appended to `_tool_log` and displayed in the Tool Calls panel
-
-**Dynamic file context refresh (ai_client.py):**
-- After the last tool call in each round, all project files from `file_items` are re-read from disk via `_reread_file_items()`. The `file_items` variable is reassigned so subsequent rounds see fresh content.
-- For Anthropic: the refreshed file contents are injected as a `text` block appended to the `tool_results` user message, prefixed with `[FILES UPDATED]` and an instruction not to re-read them.
-- For Gemini: refreshed file contents are appended to the last function response's `output` string as a `[SYSTEM: FILES UPDATED]` block. On the next tool round, stale `[FILES UPDATED]` blocks are stripped from history and old tool outputs are truncated to `_history_trunc_limit` characters to control token growth.
-- `_build_file_context_text(file_items)` formats the refreshed files as markdown code blocks (same format as the original context)
-- The `tool_result_send` comms log entry filters out the injected text block (only logs actual `tool_result` entries) to keep the comms panel clean
-- `file_items` flows from `aggregate.build_file_items()` → `gui.py` `self.last_file_items` → `ai_client.send(file_items=...)` → `_send_anthropic(file_items=...)` / `_send_gemini(file_items=...)`
-- System prompt updated to tell the AI: "the user's context files are automatically refreshed after every tool call, so you do NOT need to re-read files that are already provided in the <context> block"
-
-**Anthropic bug fixes applied (session history):**
-- Bug 1: SDK ContentBlock objects now converted to plain dicts via `_content_block_to_dict()` before storing in `_anthropic_history`; prevents re-serialisation failures on subsequent tool-use rounds
-- Bug 2: `_repair_anthropic_history` simplified to dict-only path since history always contains dicts
-- Bug 3: Gemini part.function_call access now guarded with `hasattr` check
-- Bug 4: Anthropic `b.type == "tool_use"` changed to `getattr(b, "type", None) == "tool_use"` for safe access during response processing
-
-**Comms Log (ai_client.py):**
-- `_comms_log: list[dict]` accumulates every API interaction during a session
-- `_append_comms(direction, kind, payload)` called at each boundary: OUT/request before sending, IN/response after each model reply, OUT/tool_call before executing, IN/tool_result after executing, OUT/tool_result_send when returning results to the model
-- Entry fields: `ts` (HH:MM:SS), `direction` (OUT/IN), `kind`, `provider`, `model`, `payload` (dict)
-- Anthropic responses also include `usage` (input_tokens, output_tokens, cache_creation_input_tokens, cache_read_input_tokens) and `stop_reason` in payload
-- `get_comms_log()` returns a snapshot; `clear_comms_log()` empties it
-- `comms_log_callback` (injected by gui.py) is called from the background thread with each new entry; gui queues entries in `_pending_comms` (lock-protected) and flushes them to the DPG panel each render frame
-- `COMMS_CLAMP_CHARS = 300` in gui.py governs the display cutoff for heavy text fields
-
-**Comms History panel — rich structured rendering (gui.py):**
-
-Rather than showing raw JSON, each comms entry is rendered using a kind-specific renderer function. Unknown kinds fall back to a generic key/value layout.
-
-Colour maps:
-- Direction: OUT = blue-ish `(100,200,255)`, IN = green-ish `(140,255,160)`
-- Kind: request=gold, response=light-green, tool_call=orange, tool_result=light-blue, tool_result_send=lavender
-- Labels: grey `(180,180,180)`; values: near-white `(220,220,220)`; dict keys/indices: `(140,200,255)`; numbers/token counts: `(180,255,180)`; sub-headers: `(220,200,120)`
-
-Helper functions:
-- `_add_text_field(parent, label, value)` — labelled text; strings longer than `COMMS_CLAMP_CHARS` render as an 80px readonly scrollable `input_text`; shorter strings render as `add_text`
-- `_add_kv_row(parent, key, val)` — single horizontal key: value row
-- `_render_usage(parent, usage)` — renders Anthropic token usage dict in a fixed display order (input → cache_read → cache_creation → output)
-- `_render_tool_calls_list(parent, tool_calls)` — iterates tool call list, showing name, id, and all args via `_add_text_field`
-
-Kind-specific renderers (in `_KIND_RENDERERS` dict, dispatched by `_render_comms_entry`):
-- `_render_payload_request` — shows `message` field via `_add_text_field`
-- `_render_payload_response` — shows round, stop_reason (orange), text, tool_calls list, usage block
-- `_render_payload_tool_call` — shows name, optional id, script via `_add_text_field`
-- `_render_payload_tool_result` — shows name, optional id, output via `_add_text_field`
-- `_render_payload_tool_result_send` — iterates results list, shows tool_use_id and content per result
-- `_render_payload_generic` — fallback for unknown kinds; renders all keys, using `_add_text_field` for keys in `_HEAVY_KEYS`, `_add_kv_row` for others; dicts/lists are JSON-serialised
-
-Entry layout: index + timestamp + direction + kind + provider/model header row, then payload rendered by the appropriate function, then a separator line.
-
-**Session Logger (session_logger.py):**
-- `open_session()` called once at GUI startup; creates `logs/` and `scripts/generated/` directories; opens `logs/comms_<ts>.log` and `logs/toolcalls_<ts>.log` (line-buffered)
-- `log_comms(entry)` appends each comms entry as a JSON-L line to the comms log; called from `App._on_comms_entry` (background thread); thread-safe via GIL + line buffering
-- `log_tool_call(script, result, script_path)` writes the script to `scripts/generated/<ts>_<seq:04d>.ps1` and appends a markdown record to the toolcalls log without the script body (just the file path + result); uses a `threading.Lock` for the sequence counter
-- `close_session()` flushes and closes both file handles; called just before `dpg.destroy_context()`
-
-**Anthropic prompt caching:**
-- System prompt + context are combined into one string, chunked into <=120k char blocks, and sent as the `system=` parameter array. Only the LAST chunk gets `cache_control: ephemeral`, so the entire system prefix is cached as one unit.
-- Last tool in `_ANTHROPIC_TOOLS` (`run_powershell`) has `cache_control: ephemeral`; this means the tools prefix is cached together with the system prefix after the first request.
-- The user message is sent as a plain `[{"type": "text", "text": user_message}]` block with NO cache_control. The context lives in `system=`, not in the first user message.
-- The tools list is built once per session via `_get_anthropic_tools()` and reused across all API calls within the tool loop, avoiding redundant Python-side reconstruction.
-- `_strip_cache_controls()` removes stale `cache_control` markers from all history entries before each API call, ensuring only the stable system/tools prefix consumes cache breakpoint slots.
-- Cache stats (creation tokens, read tokens) are surfaced in the comms log usage dict and displayed in the Comms History panel
-
-**Data flow:**
-1. GUI edits are held in `App` state (`self.files`, `self.screenshots`, `self.disc_entries`, `self.project`) and dpg widget values
-2. `_flush_to_project()` pulls all widget values into `self.project` dict (per-project data)
-3. `_flush_to_config()` pulls global settings into `self.config` dict
-4. `_do_generate()` calls both flush methods, saves both files, calls `project_manager.flat_config(self.project, disc_name)` to produce a dict for `aggregate.run()`, which writes the md and returns `(markdown_str, path, file_items)`
-5. `cb_generate_send()` calls `_do_generate()` then threads a call to `ai_client.send(md, message, base_dir)`
-6. `ai_client.send()` prepends the md as a `<context>` block to the user message and sends via the active provider chat session
-7. If the AI responds with tool calls, the loop handles them (with GUI confirmation) before returning the final text response
-8. Sessions are stateful within a run (chat history maintained), `Reset` clears them, the tool log, and the comms log
-
-**Config persistence:**
-- `config.toml` — global only: `[ai]` provider+model, `[theme]` palette+font+scale, `[projects]` paths array + active path
-- `<project>.toml` — per-project: output, files, screenshots, discussion (roles, active discussion name, all named discussions with their history+metadata)
-- On every send and save, both files are written
-- On clean exit, `run()` calls `_flush_to_project()`, `_save_active_project()`, `_flush_to_config()`, `save_config()` before destroying context
-
-**Threading model:**
-- DPG render loop runs on the main thread
-- AI sends and model fetches run on daemon background threads
-- `_pending_dialog` (guarded by a `threading.Lock`) is set by the background thread and consumed by the render loop each frame, calling `dialog.show()` on the main thread
-- `dialog.wait()` blocks the background thread on a `threading.Event` until the user acts
-- `_pending_comms` (guarded by a separate `threading.Lock`) is populated by `_on_comms_entry` (background thread) and drained by `_flush_pending_comms()` each render frame (main thread)
-
-**Provider error handling:**
-- `ProviderError(kind, provider, original)` wraps upstream API exceptions with a classified `kind`: quota, rate_limit, auth, balance, network, unknown
-- `_classify_anthropic_error` and `_classify_gemini_error` inspect exception types and status codes/message bodies to assign the kind
-- `ui_message()` returns a human-readable label for display in the Response panel
-
-**MCP file tools (mcp_client.py + ai_client.py):**
-- Four read-only tools exposed to the AI as native function/tool declarations: `read_file`, `list_directory`, `search_files`, `get_file_summary`
-- Access control: `mcp_client.configure(file_items, extra_base_dirs)` is called before each send; builds an allowlist of resolved absolute paths from the project's `file_items` plus the `base_dir`; any path that is not explicitly in the list or not under one of the allowed directories returns `ACCESS DENIED`
-- `mcp_client.dispatch(tool_name, tool_input)` is the single dispatch entry point used by both Anthropic and Gemini tool-use loops
-- Anthropic: MCP tools appear before `run_powershell` in the tools list (no `cache_control` on them; only `run_powershell` carries `cache_control: ephemeral`)
-- Gemini: MCP tools are included in the `FunctionDeclaration` list alongside `run_powershell`
-- `get_file_summary` uses `summarize.summarise_file()` — same heuristic used for the initial `<context>` block, so the AI gets the same compact structural view it already knows
-- `list_directory` sorts dirs before files; shows name, type, and size
-- `search_files` uses `Path.glob()` with the caller-supplied pattern (supports `**/*.py` style)
-- `read_file` returns raw UTF-8 text; errors (not found, access denied, decode error) are returned as error strings rather than exceptions, so the AI sees them as tool results
-- `summarize.py` heuristics: `.py` → AST imports + ALL_CAPS constants + classes+methods + top-level functions; `.toml` → table headers + top-level keys; `.md` → h1–h3 headings with indentation; all others → line count + first 8 lines preview
-- Comms log: MCP tool calls log `OUT/tool_call` with `{"name": ..., "args": {...}}` and `IN/tool_result` with `{"name": ..., "output": ...}`; rendered in the Comms History panel via `_render_payload_tool_call` (shows each arg key/value) and `_render_payload_tool_result` (shows output)
-
-**Known extension points:**
-- Add more providers by adding a section to `credentials.toml`, a `_list_*` and `_send_*` function in `ai_client.py`, and the provider name to the `PROVIDERS` list in `gui.py`
-- Discussion history excerpts could be individually toggleable for inclusion in the generated md
-- `MAX_TOOL_ROUNDS` in `ai_client.py` caps agentic loops at 10 rounds; adjustable
-- `COMMS_CLAMP_CHARS` in `gui.py` controls the character threshold for clamping heavy payload fields in the Comms History panel
-- Additional project metadata (description, tags, created date) could be added to `[project]` in the per-project toml
-
-### Gemini Context Management
-- Gemini uses explicit caching via `client.caches.create()` to store the `system_instruction` + tools as an immutable cached prefix with a 1-hour TTL. The cache is created once per chat session.
-- When context changes (detected via `md_content` hash), the old cache is deleted, a new cache is created, and chat history is migrated to a fresh chat session pointing at the new cache.
-- If cache creation fails (e.g., content is under the minimum token threshold — 1024 for Flash, 4096 for Pro), the system falls back to inline `system_instruction` in the chat config. Implicit caching may still provide cost savings in this case.
-- The `<context>` block lives inside `system_instruction`, NOT in user messages, preventing history bloat across turns.
-- On cleanup/exit, active caches are deleted via `ai_client.cleanup()` to prevent orphaned billing.
-
-### Latest Changes
-- Removed `Config` panel from the GUI to streamline per-project configuration.
-- `output_dir` was moved into the Projects panel.
-- `auto_add_history` was moved to the Discussion History panel.
-- `namespace` is no longer a configurable field; `aggregate.py` automatically uses the active project's `name` property.
-
-### UI / Visual Updates
-- The success blink notification on the response text box is now dimmer and more transparent to be less visually jarring.
-- Added a new floating **Last Script Output** popup window. This window automatically displays and blinks blue whenever the AI executes a PowerShell tool, showing both the executed script and its result in real-time.
-
-
-## Recent Changes (Text Viewer Maximization)
-- **Global Text Viewer (gui.py)**: Added a dedicated, large popup window (win_text_viewer) to allow reading and scrolling through large, dense text blocks without feeling cramped.
-- **Comms History**: Every multi-line text field in the comms log now has a [+] button next to its label that opens the text in the Global Text Viewer.
-- **Tool Log History**: Added [+ Script] and [+ Output] buttons next to each logged tool call to easily maximize and read the full executed scripts and raw tool outputs.
-- **Last Script Output Popup**: Expanded the default size of the popup (now 800x600) and gave the input script panel more vertical space to prevent it from feeling 'scrunched'. Added [+ Maximize] buttons for both the script and the output sections to inspect them in full detail.
-- **Confirm Dialog**: The script confirmation modal now has a [+ Maximize] button so you can read large generated scripts in full-screen before approving them.
-
-## UI Enhancements (2026-02-21)
-
-### Global Word-Wrap
-
-A new **Word-Wrap** checkbox has been added to the **Projects** panel. This setting is saved per-project in its .toml file.
-
-- When **enabled** (default), long text in read-only panels (like the main Response window, Tool Call outputs, and Comms History) will wrap to fit the panel width.
-- When **disabled**, text will not wrap, and a horizontal scrollbar will appear for oversized content.
-
-This allows you to choose the best viewing mode for either prose or wide code blocks.
-
-### Maximizable Discussion Entries
-
-Each entry in the **Discussion History** now features a [+ Max] button. Clicking this button opens the full text of that entry in the large **Text Viewer** popup, making it easy to read or copy large blocks of text from the conversation history without being constrained by the small input box.
-\n\n## Multi-Viewport & Docking\nThe application now supports Dear PyGui Viewport Docking. Windows can be dragged outside the main application area or docked together. A global 'Windows' menu in the viewport menu bar allows you to reopen any closed panels.
-
-## Extensive Documentation (2026-02-22)
-
-Documentation has been completely rewritten matching the strict, structural format of `VEFontCache-Odin`. 
-- `docs/guide_architecture.md`: Details the Python implementation algorithms, queue management for UI rendering, the specific AST heuristics used for context aggregation, and the distinct algorithms for trimming Anthropic history vs Gemini state caching.
-- `docs/Readme.md`: The core interface manual.
-- `docs/guide_tools.md`: Security architecture for `_is_allowed` paths and definitions of the read-only vs destructive tool pipeline.
-
-## Branch Analysis: master vs not_sure (2026-02-22)
-
-### Summary
-The `not_sure` branch introduces a static/dynamic context split in the `send()` API signature, separating files+screenshots (cacheable, stable) from discussion history (changes every turn). This improves cache hit rates for both Anthropic and Gemini.
-
-### Current master branch API correctness
-- **Anthropic**: Correct. System blocks with cache_control, SDK content block serialisation, history repair, stale file refresh stripping all work properly.
-- **Gemini**: Correct after `patch_gemini_history.py` was applied. Uses `_get_gemini_history_list()` for safe SDK access, drops history in pairs to maintain alternating roles, explicit caching via `caches.create()`.
-
-### not_sure branch improvements
-- **Anthropic**: Puts discussion history as a separate uncached system block after the cached static context. Better cache hit rates when discussion changes between turns.
-- **Gemini**: Wraps discussion in `<discussion>` tags in user messages and strips old ones from history via regex. Prevents discussion duplication across turns.
-
-### not_sure branch bugs (not merged from master)
-- Uses `_gemini_chat.history` directly instead of `_get_gemini_history_list()` — will crash on newer google-genai SDK versions where `.history` was removed.
-- Missing the pair-wise history dropping fix (drops single messages, breaking Gemini's alternating role requirement).
-
-### Recommended merge path
-Cherry-pick the static/dynamic split from `not_sure` into `master` while keeping master's SDK safety fixes (`_get_gemini_history_list`, pair-wise dropping, `_content_block_to_dict`).

Readme.md

@@ -1,45 +1,136 @@
-# Manual Slop
+# Sloppy
 
-Vibe coding.. but more manual
+![img](./gallery/splash.png)
 
-![img](./gallery/python_2026-02-21_23-37-29.png)
+A GUI orchestrator for local LLM-driven coding sessions. Manual Slop bridges high-latency AI reasoning with a low-latency ImGui render loop via a thread-safe asynchronous pipeline, ensuring every AI-generated payload passes through a human-auditable gate before execution.
 
-This tool is designed to work as an auxiliary assistant that natively interacts with your codebase via PowerShell and MCP-like file tools, supporting both Anthropic and Gemini APIs.
+**Tech Stack**: Python 3.11+, Dear PyGui / ImGui, FastAPI, Uvicorn
+**Providers**: Gemini API, Anthropic API, DeepSeek, Gemini CLI (headless)
+**Platform**: Windows (PowerShell) — single developer, local use
 
-Features:
+![img](./gallery/python_2026-03-01_23-45-34.png)
 
-* Multi-provider support (Anthropic & Gemini).
-* Multi-project workspace management via TOML configuration.
-* Rich discussion history with branching and timestamps.
-* Real-time file context aggregation and summarization.
-* Integrated tool execution:
-  * PowerShell scripting for file modifications.
-  * MCP-like filesystem tools (read, list, search, summarize).
-  * Web search and URL fetching.
-* Extensive UI features:
-  * Word-wrap toggles.
-  * Popup text viewers for large script/output inspection.
-  * Color theming and UI scaling.
+---
+
+## Architecture at a Glance
+
+Four thread domains operate concurrently: the ImGui main loop, an asyncio worker for AI calls, a `HookServer` (HTTP on `:8999`) for external automation, and transient threads for model fetching. Background threads never write GUI state directly — they serialize task dicts into lock-guarded lists that the main thread drains once per frame ([details](./docs/guide_architecture.md#the-task-pipeline-producer-consumer-synchronization)).
+
+The **Execution Clutch** suspends the AI execution thread on a `threading.Condition` when a destructive action (PowerShell script, sub-agent spawn) is requested. The GUI renders a modal where the user can read, edit, or reject the payload. On approval, the condition is signaled and execution resumes ([details](./docs/guide_architecture.md#the-execution-clutch-human-in-the-loop)).
+
+The **MMA (Multi-Model Agent)** system decomposes epics into tracks, tracks into DAG-ordered tickets, and executes each ticket with a stateless Tier 3 worker that starts from `ai_client.reset_session()` — no conversational bleed between tickets ([details](./docs/guide_mma.md)).
+
+---
 
 ## Documentation
 
-* [docs/Readme.md](docs/Readme.md) for the interface and usage guide
-* [docs/guide_tools.md](docs/guide_tools.md) for information on the AI tooling capabilities
-* [docs/guide_architecture.md](docs/guide_architecture.md) for an in-depth breakdown of the codebase architecture
+| Guide | Scope |
+|---|---|
+| [Architecture](./docs/guide_architecture.md) | Threading model, event system, AI client multi-provider architecture, HITL mechanism, comms logging |
+| [Tools & IPC](./docs/guide_tools.md) | MCP Bridge security model, all 26 native tools, Hook API endpoints, ApiHookClient reference, shell runner |
+| [MMA Orchestration](./docs/guide_mma.md) | 4-tier hierarchy, Ticket/Track data structures, DAG engine, ConductorEngine execution loop, worker lifecycle |
+| [Simulations](./docs/guide_simulations.md) | `live_gui` fixture, Puppeteer pattern, mock provider, visual verification patterns, ASTParser / summarizer |
 
-## Instructions
+---
 
-1. Make a credentials.toml in the immediate directory of your clone:
+## Module Map
+
+Core implementation resides in the `src/` directory.
+
+| File | Role |
+|---|---|
+| `src/gui_2.py` | Primary ImGui interface — App class, frame-sync, HITL dialogs |
+| `src/ai_client.py` | Multi-provider LLM abstraction (Gemini, Anthropic, DeepSeek, Gemini CLI) |
+| `src/mcp_client.py` | 26 MCP tools with filesystem sandboxing and tool dispatch |
+| `src/api_hooks.py` | HookServer — REST API for external automation on `:8999` |
+| `src/api_hook_client.py` | Python client for the Hook API (used by tests and external tooling) |
+| `src/multi_agent_conductor.py` | ConductorEngine — Tier 2 orchestration loop with DAG execution |
+| `src/conductor_tech_lead.py` | Tier 2 ticket generation from track briefs |
+| `src/dag_engine.py` | TrackDAG (dependency graph) + ExecutionEngine (tick-based state machine) |
+| `src/models.py` | Ticket, Track, WorkerContext dataclasses |
+| `src/events.py` | EventEmitter, AsyncEventQueue, UserRequestEvent |
+| `src/project_manager.py` | TOML config persistence, discussion management, track state |
+| `src/session_logger.py` | JSON-L + markdown audit trails (comms, tools, CLI, hooks) |
+| `src/shell_runner.py` | PowerShell execution with timeout, env config, QA callback |
+| `src/file_cache.py` | ASTParser (tree-sitter) — skeleton and curated views |
+| `src/summarize.py` | Heuristic file summaries (imports, classes, functions) |
+| `src/outline_tool.py` | Hierarchical code outline via stdlib `ast` |
+
+---
+
+## Setup
+
+### Prerequisites
+
+- Python 3.11+
+- [`uv`](https://github.com/astral-sh/uv) for package management
+
+### Installation
+
+```powershell
+git clone <repo>
+cd manual_slop
+uv sync
+```
+
+### Credentials
+
+Configure in `credentials.toml`:
 
 ```toml
 [gemini]
-api_key = "****"
+api_key = "YOUR_KEY"
+
 [anthropic]
-api_key = "****"
+api_key = "YOUR_KEY"
+
+[deepseek]
+api_key = "YOUR_KEY"
 ```
 
-2. Have fun. This is experiemntal slop.
+### Running
 
-```ps1
-uv run .\gui.py
+```powershell
+uv run sloppy.py                        # Normal mode
+uv run sloppy.py --enable-test-hooks    # With Hook API on :8999
+```
+
+### Running Tests
+
+```powershell
+uv run pytest tests/ -v
+```
+
+> **Note:** See the [Structural Testing Contract](./docs/guide_simulations.md#structural-testing-contract) for rules regarding mock patching, `live_gui` standard usage, and artifact isolation (logs are generated in `tests/logs/` and `tests/artifacts/`).
+
+---
+
+## Project Configuration
+
+Projects are stored as `<name>.toml` files. The discussion history is split into a sibling `<name>_history.toml` to keep the main config lean.
+
+```toml
+[project]
+name = "my_project"
+git_dir = "./my_repo"
+system_prompt = ""
+
+[files]
+base_dir = "./my_repo"
+paths = ["src/**/*.py", "README.md"]
+
+[screenshots]
+base_dir = "./my_repo"
+paths = []
+
+[output]
+output_dir = "./md_gen"
+
+[gemini_cli]
+binary_path = "gemini"
+
+[agent.tools]
+run_powershell = true
+read_file = true
+# ... 26 tool flags
 ```

TASKS.md

@@ -0,0 +1,129 @@
+# TASKS.md
+<!-- Quick-read pointer to active and planned conductor tracks -->
+<!-- Source of truth for task state is conductor/tracks/*/plan.md -->
+
+## Active Tracks
+*(none — all planned tracks queued below)*
+
+## Completed This Session
+*(See archive: strict_execution_queue_completed_20260306)*
+
+---
+
+#### 0. conductor_path_configurable_20260306
+- **Status:** Planned
+- **Priority:** CRITICAL
+- **Goal:** Eliminate hardcoded conductor paths. Make path configurable via config.toml or CONDUCTOR_DIR env var. Allow running app to use separate directory from development tracks.
+
+## Phase 3: Future Horizons (Tracks 1-20)
+*Initialized: 2026-03-06*
+
+### Architecture & Backend
+
+#### 1. true_parallel_worker_execution_20260306
+- **Status:** Planned
+- **Priority:** High
+- **Goal:** Implement true concurrency for the DAG engine. Once threading.local() is in place, the ExecutionEngine should spawn independent Tier 3 workers in parallel (e.g., 4 workers handling 4 isolated tests simultaneously). Requires strict file-locking or a Git-based diff-merging strategy to prevent AST collision.
+
+#### 2. deep_ast_context_pruning_20260306
+- **Status:** Planned
+- **Priority:** High
+- **Goal:** Before dispatching a Tier 3 worker, use tree_sitter to automatically parse the target file AST, strip out unrelated function bodies, and inject a surgically condensed skeleton into the worker prompt. Guarantees the AI only sees what it needs to edit, drastically reducing token burn.
+
+#### 3. visual_dag_ticket_editing_20260306
+- **Status:** Planned
+- **Priority:** Medium
+- **Goal:** Replace the linear ticket list in the GUI with an interactive Node Graph using ImGui Bundle node editor. Allow the user to visually drag dependency lines, split nodes, or delete tasks before clicking Execute Pipeline.
+
+#### 4. tier4_auto_patching_20260306
+- **Status:** Planned
+- **Priority:** Medium
+- **Goal:** Elevate Tier 4 from a log summarizer to an auto-patcher. When a verification test fails, Tier 4 generates a .patch file. The GUI intercepts this and presents a side-by-side Diff Viewer. The user clicks Apply Patch to instantly resume the pipeline.
+
+#### 5. native_orchestrator_20260306
+- **Status:** Planned
+- **Priority:** Low
+- **Goal:** Absorb the Conductor extension entirely into the core application. Manual Slop should natively read/write plan.md, manage the metadata.json, and orchestrate the MMA tiers in pure Python, removing the dependency on external CLI shell executions (mma_exec.py).
+
+---
+
+### GUI Overhauls & Visualizations
+
+#### 6. cost_token_analytics_20260306
+- **Status:** Planned
+- **Priority:** High
+- **Goal:** Real-time cost tracking panel displaying cost per model, session totals, and breakdown by tier. Uses existing cost_tracker.py which is implemented but has no GUI.
+
+#### 7. performance_dashboard_20260306
+- **Status:** Planned
+- **Priority:** High
+- **Goal:** Expand performance metrics panel with CPU/RAM usage, frame time, input lag with historical graphs. Uses existing performance_monitor.py which has basic metrics but no detailed visualization.
+
+#### 8. mma_multiworker_viz_20260306
+- **Status:** Planned
+- **Priority:** High
+- **Goal:** Split-view GUI for parallel worker streams per tier. Visualize multiple concurrent workers with individual status, output tabs, and resource usage. Enable kill/restart per worker.
+
+#### 9. cache_analytics_20260306
+- **Status:** Planned
+- **Priority:** Medium
+- **Goal:** Gemini cache hit/miss visualization, memory usage, TTL status display. Uses existing ai_client.get_gemini_cache_stats() which is not displayed in GUI.
+
+#### 10. tool_usage_analytics_20260306
+- **Status:** Planned
+- **Priority:** Medium
+- **Goal:** Analytics panel showing most-used tools, average execution time, and failure rates. Uses existing tool_log_callback data.
+
+#### 11. session_insights_20260306
+- **Status:** Planned
+- **Priority:** Medium
+- **Goal:** Token usage over time, cost projections, session summary with efficiency scores. Visualize session_logger data.
+
+#### 12. track_progress_viz_20260306
+- **Status:** Planned
+- **Priority:** Medium
+- **Goal:** Progress bars and percentage completion for active tracks and tickets. Better visualization of DAG execution state.
+
+#### 13. manual_skeleton_injection_20260306
+- **Status:** Planned
+- **Priority:** Medium
+- **Goal:** Add UI controls to manually flag files for skeleton injection in discussions. Allow agent to request full file reads or specific def/class definitions on-demand.
+
+#### 14. on_demand_def_lookup_20260306
+- **Status:** Planned
+- **Priority:** Medium
+- **Goal:** Add ability for agent to request specific class/function definitions during discussion. User can @mention a symbol and get its full definition inline.
+
+---
+
+### Manual UX Controls
+
+#### 15. ticket_queue_mgmt_20260306
+- **Status:** Planned
+- **Priority:** High
+- **Goal:** Allow user to manually reorder, prioritize, or requeue tickets in the DAG. Add drag-drop reordering, priority tags, and bulk selection.
+
+#### 16. kill_abort_workers_20260306
+- **Status:** Planned
+- **Priority:** High
+- **Goal:** Add ability to kill/abort a running Tier 3 worker mid-execution. Currently workers run to completion; add cancel button.
+
+#### 17. manual_block_control_20260306
+- **Status:** Planned
+- **Priority:** Medium
+- **Goal:** Allow user to manually block or unblock tickets with custom reasons. Currently blocked tickets rely on dependency resolution; add manual override.
+
+#### 18. pipeline_pause_resume_20260306
+- **Status:** Planned
+- **Priority:** Medium
+- **Goal:** Add global pause/resume for the entire DAG execution pipeline. Allow user to freeze all worker activity and resume later.
+
+#### 19. per_ticket_model_20260306
+- **Status:** Planned
+- **Priority:** Low
+- **Goal:** Allow user to manually select which model to use for a specific ticket, overriding the default tier model.
+
+#### 20. manual_ux_validation_20260302
+- **Status:** Planned
+- **Priority:** Medium
+- **Goal:** Interactive human-in-the-loop track to review and adjust GUI UX, animations, popups, and layout structures.

aggregate.py

@@ -1,193 +0,0 @@
-# aggregate.py
-"""
-Note(Gemini):
-This module orchestrates the construction of the final Markdown context string. 
-Instead of sending every file to the AI raw (which blows up tokens), this uses a pipeline:
-1. Resolve paths (handles globs and absolute paths).
-2. Build file items (raw content).
-3. If 'summary_only' is true (which is the default behavior now), it pipes the files through 
-   summarize.py to generate a compacted view.
-   
-This is essential for keeping prompt tokens low while giving the AI enough structural info 
-to use the MCP tools to fetch only what it needs.
-"""
-import tomllib
-import re
-import glob
-from pathlib import Path, PureWindowsPath
-import summarize
-
-def find_next_increment(output_dir: Path, namespace: str) -> int:
-    pattern = re.compile(rf"^{re.escape(namespace)}_(\d+)\.md$")
-    max_num = 0
-    for f in output_dir.iterdir():
-        if f.is_file():
-            match = pattern.match(f.name)
-            if match:
-                max_num = max(max_num, int(match.group(1)))
-    return max_num + 1
-
-def is_absolute_with_drive(entry: str) -> bool:
-    try:
-        p = PureWindowsPath(entry)
-        return p.drive != ""
-    except Exception:
-        return False
-
-def resolve_paths(base_dir: Path, entry: str) -> list[Path]:
-    has_drive = is_absolute_with_drive(entry)
-    is_wildcard = "*" in entry
-    if is_wildcard:
-        root = Path(entry) if has_drive else base_dir / entry
-        matches = [Path(p) for p in glob.glob(str(root), recursive=True) if Path(p).is_file()]
-        return sorted(matches)
-    else:
-        if has_drive:
-            return [Path(entry)]
-        return [(base_dir / entry).resolve()]
-
-def build_discussion_section(history: list[str]) -> str:
-    sections = []
-    for i, paste in enumerate(history, start=1):
-        sections.append(f"### Discussion Excerpt {i}\n\n{paste.strip()}")
-    return "\n\n---\n\n".join(sections)
-
-def build_files_section(base_dir: Path, files: list[str]) -> str:
-    sections = []
-    for entry in files:
-        paths = resolve_paths(base_dir, entry)
-        if not paths:
-            sections.append(f"### `{entry}`\n\n```text\nERROR: no files matched: {entry}\n```")
-            continue
-        for path in paths:
-            suffix = path.suffix.lstrip(".")
-            lang = suffix if suffix else "text"
-            try:
-                content = path.read_text(encoding="utf-8")
-            except FileNotFoundError:
-                content = f"ERROR: file not found: {path}"
-            except Exception as e:
-                content = f"ERROR: {e}"
-            original = entry if "*" not in entry else str(path)
-            sections.append(f"### `{original}`\n\n```{lang}\n{content}\n```")
-    return "\n\n---\n\n".join(sections)
-
-def build_screenshots_section(base_dir: Path, screenshots: list[str]) -> str:
-    sections = []
-    for entry in screenshots:
-        paths = resolve_paths(base_dir, entry)
-        if not paths:
-            sections.append(f"### `{entry}`\n\n_ERROR: no files matched: {entry}_")
-            continue
-        for path in paths:
-            original = entry if "*" not in entry else str(path)
-            if not path.exists():
-                sections.append(f"### `{original}`\n\n_ERROR: file not found: {path}_")
-                continue
-            sections.append(f"### `{original}`\n\n![{path.name}]({path.as_posix()})")
-    return "\n\n---\n\n".join(sections)
-
-
-def build_file_items(base_dir: Path, files: list[str]) -> list[dict]:
-    """
-    Return a list of dicts describing each file, for use by ai_client when it
-    wants to upload individual files rather than inline everything as markdown.
-
-    Each dict has:
-        path     : Path  (resolved absolute path)
-        entry    : str   (original config entry string)
-        content  : str   (file text, or error string)
-        error    : bool
-    """
-    items = []
-    for entry in files:
-        paths = resolve_paths(base_dir, entry)
-        if not paths:
-            items.append({"path": None, "entry": entry, "content": f"ERROR: no files matched: {entry}", "error": True})
-            continue
-        for path in paths:
-            try:
-                content = path.read_text(encoding="utf-8")
-                error = False
-            except FileNotFoundError:
-                content = f"ERROR: file not found: {path}"
-                error = True
-            except Exception as e:
-                content = f"ERROR: {e}"
-                error = True
-            items.append({"path": path, "entry": entry, "content": content, "error": error})
-    return items
-
-def build_summary_section(base_dir: Path, files: list[str]) -> str:
-    """
-    Build a compact summary section using summarize.py — one short block per file.
-    Used as the initial <context> block instead of full file contents.
-    """
-    items = build_file_items(base_dir, files)
-    return summarize.build_summary_markdown(items)
-
-def build_static_markdown(base_dir: Path, files: list[str], screenshot_base_dir: Path, screenshots: list[str], summary_only: bool = False) -> str:
-    """Build the static (cacheable) portion of the context: files + screenshots."""
-    parts = []
-    if files:
-        if summary_only:
-            parts.append("## Files (Summary)\n\n" + build_summary_section(base_dir, files))
-        else:
-            parts.append("## Files\n\n" + build_files_section(base_dir, files))
-    if screenshots:
-        parts.append("## Screenshots\n\n" + build_screenshots_section(screenshot_base_dir, screenshots))
-    return "\n\n---\n\n".join(parts)
-
-def build_dynamic_markdown(history: list[str]) -> str:
-    """Build the dynamic (changes every turn) portion: discussion history."""
-    if history:
-        return "## Discussion History\n\n" + build_discussion_section(history)
-    return ""
-
-def build_markdown(base_dir: Path, files: list[str], screenshot_base_dir: Path, screenshots: list[str], history: list[str], summary_only: bool = False) -> str:
-    parts = []
-    # STATIC PREFIX: Files and Screenshots must go first to maximize Cache Hits
-    if files:
-        if summary_only:
-            parts.append("## Files (Summary)\n\n" + build_summary_section(base_dir, files))
-        else:
-            parts.append("## Files\n\n" + build_files_section(base_dir, files))
-    if screenshots:
-        parts.append("## Screenshots\n\n" + build_screenshots_section(screenshot_base_dir, screenshots))
-    # DYNAMIC SUFFIX: History changes every turn, must go last
-    if history:
-        parts.append("## Discussion History\n\n" + build_discussion_section(history))
-    return "\n\n---\n\n".join(parts)
-
-def run(config: dict) -> tuple[str, Path]:
-    namespace = config.get("project", {}).get("name")
-    if not namespace:
-        namespace = config.get("output", {}).get("namespace", "project")
-    output_dir = Path(config["output"]["output_dir"])
-    base_dir = Path(config["files"]["base_dir"])
-    files = config["files"].get("paths", [])
-    screenshot_base_dir = Path(config.get("screenshots", {}).get("base_dir", "."))
-    screenshots = config.get("screenshots", {}).get("paths", [])
-    history = config.get("discussion", {}).get("history", [])
-
-    output_dir.mkdir(parents=True, exist_ok=True)
-    increment = find_next_increment(output_dir, namespace)
-    output_file = output_dir / f"{namespace}_{increment:03d}.md"
-    # Build static (files+screenshots) and dynamic (discussion) portions separately for better caching
-    static_md = build_static_markdown(base_dir, files, screenshot_base_dir, screenshots, summary_only=False)
-    dynamic_md = build_dynamic_markdown(history)
-    # Write combined markdown to disk for archival
-    markdown = f"{static_md}\n\n---\n\n{dynamic_md}" if static_md and dynamic_md else static_md or dynamic_md
-    output_file.write_text(markdown, encoding="utf-8")
-    file_items = build_file_items(base_dir, files)
-    return static_md, dynamic_md, output_file, file_items
-
-def main():
-    with open("config.toml", "rb") as f:
-        import tomllib
-        config = tomllib.load(f)
-    static_md, dynamic_md, output_file, _ = run(config)
-    print(f"Written: {output_file}")
-
-if __name__ == "__main__":
-    main()

conductor/archive/api_hooks_verification_20260223/index.md

@@ -0,0 +1,5 @@
+# Track api_hooks_verification_20260223 Context
+
+- [Specification](./spec.md)
+- [Implementation Plan](./plan.md)
+- [Metadata](./metadata.json)

conductor/archive/api_hooks_verification_20260223/metadata.json

@@ -0,0 +1,8 @@
+{
+  "track_id": "api_hooks_verification_20260223",
+  "type": "feature",
+  "status": "new",
+  "created_at": "2026-02-23T17:46:51Z",
+  "updated_at": "2026-02-23T17:46:51Z",
+  "description": "Update conductor to properly utilize the new api hooks for automated testing & verification of track implementation features without the need of user intervention."
+}

conductor/archive/api_hooks_verification_20260223/plan.md

@@ -0,0 +1,19 @@
+# Implementation Plan: Integrate API Hooks for Automated Track Verification
+
+## Phase 1: Update Workflow Definition [checkpoint: f17c9e3]
+- [x] Task: Modify `conductor/workflow.md` to reflect the new automated verification process. [2ec1ecf]
+    - [ ] Sub-task: Update the "Phase Completion Verification and Checkpointing Protocol" section to replace manual verification steps with a description of the automated API hook process.
+    - [ ] Sub-task: Ensure the updated workflow clearly states that the agent will announce the automated test, execute it, and then present the results (success or failure) to the user.
+- [ ] Task: Conductor - User Manual Verification 'Phase 1: Update Workflow Definition' (Protocol in workflow.md)
+
+## Phase 2: Implement Automated Verification Logic [checkpoint: b575dcd]
+- [x] Task: Develop the client-side logic for communicating with the API hook server. [f4a9ff8]
+    - [ ] Sub-task: Write failing unit tests for a new `ApiHookClient` that can send requests to the IPC server.
+    - [ ] Sub-task: Implement the `ApiHookClient` to make the tests pass.
+- [x] Task: Integrate the `ApiHookClient` into the Conductor agent's workflow. [c7c8b89]
+    - [ ] Sub-task: Write failing integration tests to ensure the Conductor's phase completion logic calls the `ApiHookClient`.
+    - [ ] Sub-task: Modify the workflow implementation to use the `ApiHookClient` for verification.
+- [x] Task: Implement result handling and user feedback. [94b4f38]
+    - [ ] Sub-task: Write failing tests for handling success, failure, and server-unavailable scenarios.
+    - [ ] Sub-task: Implement the logic to log results, present them to the user, and halt the workflow on failure.
+- [ ] Task: Conductor - User Manual Verification 'Phase 2: Implement Automated Verification Logic' (Protocol in workflow.md)

conductor/archive/api_hooks_verification_20260223/spec.md

@@ -0,0 +1,21 @@
+# Specification: Integrate API Hooks for Automated Track Verification
+
+## Overview
+This track focuses on integrating the existing, previously implemented API hooks (from track `test_hooks_20260223`) into the Conductor workflow. The primary goal is to automate the verification steps within the "Phase Completion Verification and Checkpointing Protocol", reducing the need for manual user intervention and enabling a more streamlined, automated development process.
+
+## Functional Requirements
+- **Workflow Integration:** The `workflow.md` document, specifically the "Phase Completion Verification and Checkpointing Protocol," must be updated to replace manual verification steps with automated checks using the API hooks.
+- **IPC Communication:** The updated workflow will communicate with the application's backend via the established IPC server to trigger verification tasks.
+- **Result Handling:**
+    - All results from the API hook verifications must be logged for auditing and debugging purposes.
+    - Upon successful verification, the Conductor agent will proceed with the workflow as it currently does after a successful manual check.
+    - Upon failure, the agent will halt, present the failure logs to the user, and await further instructions.
+- **User Interaction Model:** The system will transition from asking the user to perform a manual test to informing the user that an automated test is running, and then presenting the results.
+
+## Non-Functional Requirements
+- **Resilience:** The Conductor agent must handle cases where the API hook server is unavailable or a hook call fails unexpectedly, without crashing or entering an unrecoverable state.
+- **Transparency:** All interactions with the API hooks must be clearly logged, making the automated process easy to monitor and debug.
+
+## Out of Scope
+- **Modifying API Hooks:** This track will not alter the existing API hooks, the IPC server, or the backend implementation. The focus is solely on the client-side integration within the Conductor agent's workflow.
+- **Changes to Manual Overrides:** Users will retain the ability to manually intervene or bypass automated checks if necessary.

conductor/archive/api_metrics_20260223/index.md

@@ -0,0 +1,5 @@
+# Track api_metrics_20260223 Context
+
+- [Specification](./spec.md)
+- [Implementation Plan](./plan.md)
+- [Metadata](./metadata.json)

conductor/archive/api_metrics_20260223/metadata.json

@@ -0,0 +1,8 @@
+{
+  "track_id": "api_metrics_20260223",
+  "type": "feature",
+  "status": "new",
+  "created_at": "2026-02-23T10:00:00Z",
+  "updated_at": "2026-02-23T10:00:00Z",
+  "description": "Review vendor api usage in regards to conservative context handling"
+}

conductor/archive/api_metrics_20260223/plan.md

@@ -0,0 +1,19 @@
+# Implementation Plan
+
+## Phase 1: Metric Extraction and Logic Review [checkpoint: 2668f88]
+- [x] Task: Extract explicit cache counts and lifecycle states from Gemini SDK
+    - [x] Sub-task: Write Tests
+    - [x] Sub-task: Implement Feature
+- [x] Task: Review and expose 'history bleed' (token limit proximity) flags
+    - [x] Sub-task: Write Tests
+    - [x] Sub-task: Implement Feature
+- [x] Task: Conductor - User Manual Verification 'Phase 1: Metric Extraction and Logic Review' (Protocol in workflow.md)
+
+## Phase 2: GUI Telemetry and Plotting [checkpoint: 76582c8]
+- [x] Task: Implement token budget visualizer (e.g., Progress bars for limits) in Dear PyGui
+    - [x] Sub-task: Write Tests
+    - [x] Sub-task: Implement Feature
+- [x] Task: Implement active caches data display in Provider/Comms panel
+    - [x] Sub-task: Write Tests
+    - [x] Sub-task: Implement Feature
+- [x] Task: Conductor - User Manual Verification 'Phase 2: GUI Telemetry and Plotting' (Protocol in workflow.md)

conductor/archive/api_metrics_20260223/spec.md

@@ -0,0 +1,22 @@
+# Specification: Review vendor api usage in regards to conservative context handling
+
+## Overview
+This track aims to optimize token efficiency and transparency by reviewing and improving how vendor APIs (Gemini and Anthropic) handle conservative context pruning. The primary focus is on extracting, plotting, and exposing deep metrics to the GUI so developers can intuit how close they are to API limits (e.g., token caps, cache counts, history bleed).
+
+## Scope
+- **Gemini Hooks:** Review explicit context caching, cache invalidation, and tools declaration. 
+- **Global Orchestration:** Review global context boundaries within the main prompt lifecycle.
+- **GUI Metrics:** Expose as much metric data as possible to the user interface (e.g., plotting token usage, visual indicators for when "history bleed" occurs, displaying the number of active caches).
+
+## Functional Requirements
+- Implement extensive token and cache metric extraction from both Gemini and Anthropic API responses.
+- Expose these metrics to the Dear PyGui frontend, potentially utilizing visual plots or progress bars to indicate token budget consumption.
+- Implement tests to explicitly verify context rules, ensuring history pruning acts conservatively and predictable without data loss.
+
+## Non-Functional Requirements
+- Ensure GUI rendering of new plots or dense metrics does not block the main thread.
+- Adhere to the "Strict State Management" product guideline.
+
+## Out of Scope
+- Major feature additions unrelated to context token management or telemetry.
+- Expanding the AI's agentic capabilities (e.g., new tools).

conductor/archive/api_vendor_alignment_20260223/index.md

@@ -0,0 +1,5 @@
+# Track api_vendor_alignment_20260223 Context
+
+- [Specification](./spec.md)
+- [Implementation Plan](./plan.md)
+- [Metadata](./metadata.json)

conductor/archive/api_vendor_alignment_20260223/metadata.json

@@ -0,0 +1,8 @@
+{
+  "track_id": "api_vendor_alignment_20260223",
+  "type": "chore",
+  "status": "new",
+  "created_at": "2026-02-23T12:00:00Z",
+  "updated_at": "2026-02-23T12:00:00Z",
+  "description": "Review project codebase, documentation related to project, and make sure agenti vendor apis are being used as properly stated by offical documentation from google for gemini and anthropic for claude."
+}

conductor/archive/api_vendor_alignment_20260223/plan.md

@@ -0,0 +1,56 @@
+# Implementation Plan: API Usage Audit and Alignment
+
+## Phase 1: Research and Comprehensive Audit [checkpoint: 5ec4283]
+Identify all points of interaction with AI SDKs and compare them with latest official documentation.
+
+- [x] Task: List and categorize all AI SDK usage in the project.
+    - [x] Search for all imports of `google.genai` and `anthropic`.
+    - [x] Document specific functions and methods being called.
+- [x] Task: Research latest official documentation for `google-genai` and `anthropic` Python SDKs.
+    - [x] Verify latest patterns for Client initialization.
+    - [x] Verify latest patterns for Context/Prompt caching.
+    - [x] Verify latest patterns for Tool/Function calling.
+- [x] Task: Conductor - User Manual Verification 'Phase 1: Research and Comprehensive Audit' (Protocol in workflow.md)
+
+## Phase 2: Gemini (google-genai) Alignment [checkpoint: 842bfc4]
+Align Gemini integration with documented best practices.
+
+- [x] Task: Refactor Gemini Client and Chat initialization if needed.
+    - [x] Write Tests
+    - [x] Implement Feature
+- [x] Task: Optimize Gemini Context Caching.
+    - [x] Write Tests
+    - [x] Implement Feature
+- [x] Task: Align Gemini Tool Declaration and handling.
+    - [x] Write Tests
+    - [x] Implement Feature
+- [x] Task: Conductor - User Manual Verification 'Phase 2: Gemini (google-genai) Alignment' (Protocol in workflow.md)
+
+## Phase 3: Anthropic Alignment [checkpoint: f0eb538]
+Align Anthropic integration with documented best practices.
+
+- [x] Task: Refactor Anthropic Client and Message creation if needed.
+    - [x] Write Tests
+    - [x] Implement Feature
+- [x] Task: Optimize Anthropic Prompt Caching (`cache_control`).
+    - [x] Write Tests
+    - [x] Implement Feature
+- [x] Task: Align Anthropic Tool Declaration and handling.
+    - [x] Write Tests
+    - [x] Implement Feature
+- [x] Task: Conductor - User Manual Verification 'Phase 3: Anthropic Alignment' (Protocol in workflow.md)
+
+## Phase 4: History and Token Management [checkpoint: 0f9f235]
+Ensure accurate token estimation and robust history handling.
+
+- [x] Task: Review and align token estimation logic for both providers.
+    - [x] Write Tests
+    - [x] Implement Feature
+- [x] Task: Audit message history truncation and context window management.
+    - [x] Write Tests
+    - [x] Implement Feature
+- [x] Task: Conductor - User Manual Verification 'Phase 4: History and Token Management' (Protocol in workflow.md)
+
+## Phase 5: Final Validation and Cleanup [checkpoint: e9126b4]
+- [x] Task: Perform a full test run using `run_tests.py` to ensure 100% pass rate.
+- [x] Task: Conductor - User Manual Verification 'Phase 5: Final Validation and Cleanup' (Protocol in workflow.md)

conductor/archive/api_vendor_alignment_20260223/spec.md

@@ -0,0 +1,29 @@
+# Specification: API Usage Audit and Alignment
+
+## Overview
+This track involves a comprehensive audit of the "Manual Slop" codebase to ensure that the integration with Google Gemini (`google-genai`) and Anthropic Claude (`anthropic`) SDKs aligns perfectly with their latest official documentation and best practices. The goal is to identify discrepancies, performance bottlenecks, or deprecated patterns and implement the necessary fixes.
+
+## Scope
+- **Target:** Full codebase audit, with primary focus on `ai_client.py`, `mcp_client.py`, and any other modules interacting with AI SDKs.
+- **Key Areas:**
+    - **Caching Mechanisms:** Verify Gemini context caching and Anthropic prompt caching implementation.
+    - **Tool Calling:** Audit function declarations, parameter schemas, and result handling.
+    - **History & Tokens:** Review message history management, token estimation accuracy, and context window handling.
+
+## Functional Requirements
+1. **SDK Audit:** Compare existing code patterns against the latest official Python SDK documentation for Gemini and Anthropic.
+2. **Feature Validation:**
+    - Ensure `google-genai` usage follows the latest `Client` and `types` patterns.
+    - Ensure `anthropic` usage utilizes `cache_control` correctly for optimal performance.
+3. **Discrepancy Remediation:** Implement code changes to align the implementation with documented standards.
+4. **Validation:** Execute tests to ensure that API interactions remain functional and improved.
+
+## Acceptance Criteria
+- Full audit completed for all AI SDK interactions.
+- Identified discrepancies are documented and fixed.
+- Caching, tool calling, and history management logic are verified against latest SDK standards.
+- All existing and new tests pass successfully.
+
+## Out of Scope
+- Adding support for new AI providers not already in the project.
+- Major UI refactoring unless directly required by API changes.

conductor/archive/architecture_boundary_hardening_20260302/index.md

@@ -0,0 +1,5 @@
+# Track architecture_boundary_hardening_20260302 Context
+
+- [Specification](./spec.md)
+- [Implementation Plan](./plan.md)
+- [Metadata](./metadata.json)

conductor/archive/architecture_boundary_hardening_20260302/metadata.json

@@ -0,0 +1,8 @@
+{
+  "track_id": "architecture_boundary_hardening_20260302",
+  "type": "fix",
+  "status": "new",
+  "created_at": "2026-03-02T00:00:00Z",
+  "updated_at": "2026-03-02T00:00:00Z",
+  "description": "Fix boundary leak where the native MCP file mutation tools bypass the manual_slop GUI approval dialog, and patch token leaks in the meta-tooling scripts."
+}

conductor/archive/architecture_boundary_hardening_20260302/plan.md

@@ -0,0 +1,25 @@
+# Implementation Plan: Architecture Boundary Hardening
+
+Architecture reference: [docs/guide_architecture.md](../../../docs/guide_architecture.md)
+
+---
+
+## Phase 1: Patch Context Amnesia Leak & Portability (Meta-Tooling) [checkpoint: 15536d7]
+Focus: Stop `mma_exec.py` from injecting massive full-text dependencies and remove hardcoded external paths.
+
+- [x] Task 1.1: In `scripts/mma_exec.py`, completely remove the `UNFETTERED_MODULES` constant and its associated `if dep in UNFETTERED_MODULES:` check. Ensure all imported local dependencies strictly use `generate_skeleton()`. 6875459
+- [x] Task 1.2: In `scripts/mma_exec.py` and `scripts/claude_mma_exec.py`, remove the hardcoded reference to `C:\projects\misc\setup_*.ps1`. Rely on the active environment's PATH to resolve `gemini` and `claude`, or provide an `.env` configurable override. b30f040
+
+## Phase 2: Complete MCP Tool Integration & Seal HITL Bypass (Application Core) [checkpoint: 1a65b11]
+Focus: Expose all native MCP tools in the config and GUI, and ensure mutating tools trigger user approval.
+
+- [x] Task 2.1: Update `manual_slop.toml` and `project_manager.py`'s `default_project()` to include all new tools (e.g., `set_file_slice`, `py_update_definition`, `py_set_signature`) under `[agent.tools]`. e4ccb06
+- [x] Task 2.2: Update `gui_2.py`'s settings/config panels to expose toggles for these new tools. 4b7338a
+- [x] Task 2.3: In `mcp_client.py`, define a `MUTATING_TOOLS` constant set. 1f92629
+- [x] Task 2.4: In `ai_client.py`'s provider loops (`_send_gemini`, `_send_gemini_cli`, `_send_anthropic`, `_send_deepseek`), update the tool execution logic: if `name in mcp_client.MUTATING_TOOLS`, it MUST trigger a GUI approval mechanism (like `pre_tool_callback`) before dispatching the tool. e5e35f7
+
+## Phase 3: DAG Engine Cascading Blocks (Application Core) [checkpoint: 80d79fe]
+Focus: Prevent infinite deadlocks when Tier 3 workers fail repeatedly.
+
+- [x] Task 3.1: In `dag_engine.py`, add a `cascade_blocks()` method to `TrackDAG`. This method should iterate through all `todo` tickets and if any of their dependencies are `blocked`, mark the ticket itself as `blocked`. 5b8a073
+- [x] Task 3.2: In `multi_agent_conductor.py`, update `ConductorEngine.run()`. Before calling `self.engine.tick()`, call `self.track_dag.cascade_blocks()` (or equivalent) so that blocked states propagate cleanly, allowing the `all_done` or block detection logic to exit the while loop correctly. 5b8a073

conductor/archive/architecture_boundary_hardening_20260302/spec.md

@@ -0,0 +1,28 @@
+# Track Specification: Architecture Boundary Hardening
+
+## Overview
+The `manual_slop` project sandbox provides AI meta-tooling (`mma_exec.py`, `tool_call.py`) to orchestrate its own development. When AI agents added advanced AST tools (like `set_file_slice`) to `mcp_client.py` for meta-tooling, they failed to fully integrate them into the application's GUI, config, or HITL (Human-In-The-Loop) safety models. Additionally, meta-tooling scripts are bleeding tokens and rely on non-portable hardcoded machine paths, while the internal application's state machine can deadlock.
+
+## Current State Audit
+
+1. **Incomplete MCP Tool Integration & HITL Bypass (`ai_client.py`, `gui_2.py`)**:
+   - Issue: New tools in `mcp_client.py` (e.g., `set_file_slice`, `py_update_definition`) are not exposed in the GUI or `manual_slop.toml` config `[agent.tools]`. If they were enabled, `ai_client.py` would execute them instantly without checking `pre_tool_callback`, bypassing GUI approval.
+   - *Requirement*: Expose all `mcp_client.py` tools as toggles in the GUI/Config. Ensure any mutating tool triggers a GUI approval modal before execution.
+
+2. **Token Firewall Leak in Meta-Tooling (`mma_exec.py`)**:
+   - Location: `scripts/mma_exec.py:101`.
+   - Issue: `UNFETTERED_MODULES` hardcodes `['mcp_client', 'project_manager', 'events', 'aggregate']`. If a worker targets a file that imports `mcp_client`, the script injects the full `mcp_client.py` (~450 lines) into the context instead of its skeleton, blowing out the token budget.
+
+3. **Portability Leak in Meta-Tooling Scripts**:
+   - Location: `scripts/mma_exec.py` and `scripts/claude_mma_exec.py`.
+   - Issue: Both scripts hardcode absolute external paths (`C:\projects\misc\setup_gemini.ps1` and `setup_claude.ps1`) to initialize the subprocess environment. This breaks repository portability.
+
+4. **DAG Engine Blocking Stalls (`dag_engine.py`)**:
+   - Location: `dag_engine.py` -> `get_ready_tasks()`
+   - Issue: `get_ready_tasks` requires all dependencies to be explicitly `completed`. If a task is marked `blocked`, its dependents stay `todo` forever, causing an infinite stall.
+
+## Desired State
+- All tools in `mcp_client.py` are configurable in `manual_slop.toml` and `gui_2.py`. Mutating tools must route through the GUI approval callback.
+- The `UNFETTERED_MODULES` list must be completely removed from `mma_exec.py`.
+- Meta-tooling scripts rely on standard PATH or local relative config files, not hardcoded absolute external paths.
+- The `dag_engine.py` must cascade `blocked` status to downstream tasks so the track halts cleanly.

conductor/archive/codebase_migration_20260302/index.md

@@ -0,0 +1,5 @@
+# Track codebase_migration_20260302 Context
+
+- [Specification](./spec.md)
+- [Implementation Plan](./plan.md)
+- [Metadata](./metadata.json)

conductor/archive/codebase_migration_20260302/metadata.json

@@ -0,0 +1,8 @@
+{
+  "track_id": "codebase_migration_20260302",
+  "type": "chore",
+  "status": "new",
+  "created_at": "2026-03-02T22:28:00Z",
+  "updated_at": "2026-03-02T22:28:00Z",
+  "description": "Move the codebase from the main directory to a src directory. Alleviate clutter by doing so. Remove files that are not used at all by the current application's implementation."
+}

conductor/archive/codebase_migration_20260302/plan.md

@@ -0,0 +1,23 @@
+# Implementation Plan: Codebase Migration to `src` & Cleanup (codebase_migration_20260302)
+
+## Status: COMPLETE [checkpoint: 92da972]
+
+## Phase 1: Unused File Identification & Removal
+- [x] Task: Initialize MMA Environment `activate_skill mma-orchestrator`
+- [x] Task: Audit Codebase for Dead Files (1eb9d29)
+- [x] Task: Delete Unused Files (1eb9d29)
+- [-] Task: Conductor - User Manual Verification 'Phase 1: Unused File Identification & Removal' (SKIPPED)
+
+## Phase 2: Directory Restructuring & Migration
+- [x] Task: Create `src/` Directory
+- [x] Task: Move Application Files to `src/`
+- [x] Task: Conductor - User Manual Verification 'Phase 2: Directory Restructuring & Migration' (Checkpoint: 24f385e)
+
+## Phase 3: Entry Point & Import Resolution
+- [x] Task: Create `sloppy.py` Entry Point (c102392)
+- [x] Task: Resolve Absolute and Relative Imports (c102392)
+- [x] Task: Conductor - User Manual Verification 'Phase 3: Entry Point & Import Resolution' (Checkpoint: 24f385e)
+## Phase 4: Final Validation & Documentation
+- [x] Task: Full Test Suite Validation (ea5bb4e)
+- [x] Task: Update Core Documentation (ea5bb4e)
+- [x] Task: Conductor - User Manual Verification 'Phase 4: Final Validation & Documentation' (92da972)

conductor/archive/codebase_migration_20260302/spec.md

@@ -0,0 +1,33 @@
+# Track Specification: Codebase Migration to `src` & Cleanup (codebase_migration_20260302)
+
+## Overview
+This track focuses on restructuring the codebase to alleviate clutter by moving the main implementation files from the project root into a dedicated `src/` directory. Additionally, files that are completely unused by the current implementation will be automatically identified and removed. A new clean entry point (`sloppy.py`) will be created in the root directory.
+
+## Functional Requirements
+- **Directory Restructuring**:
+  - Move all active Python implementation files (e.g., `gui_2.py`, `ai_client.py`, `mcp_client.py`, `shell_runner.py`, `project_manager.py`, `events.py`, etc.) into a new `src/` directory.
+  - Update internal imports within all moved files to reflect their new locations or ensure the Python path resolves them correctly.
+- **Root Directory Retention**:
+  - Keep configuration files (e.g., `config.toml`, `pyproject.toml`, `requirements.txt`, `.gitignore`) in the project root.
+  - Keep documentation files and directories (e.g., `Readme.md`, `BUILD.md`, `docs/`) in the project root.
+  - Keep the `tests/` and `simulation/` directories at the root level.
+- **New Entry Point**:
+  - Create a new file `sloppy.py` in the root directory.
+  - `sloppy.py` will serve as the primary entry point to launch the application (jumpstarting the underlying `gui_2.py` logic which will be moved into `src/`).
+- **Dead Code/File Removal**:
+  - Automatically identify completely unused files and scripts in the project root (e.g., legacy files, unreferenced tools).
+  - Delete the identified unused files to clean up the repository.
+
+## Non-Functional Requirements
+- Ensure all automated tests (`tests/`) and simulations (`simulation/`) continue to function perfectly without `ModuleNotFoundError`s.
+- `sloppy.py` must support existing CLI arguments (e.g., `--enable-test-hooks`).
+
+## Acceptance Criteria
+- [ ] A `src/` directory exists and contains the main application logic.
+- [ ] The root directory is clean, containing mainly configs, docs, `tests/`, `simulation/`, and `sloppy.py`.
+- [ ] `sloppy.py` successfully launches the application.
+- [ ] The full test suite runs and passes (i.e. all imports are correctly resolved).
+- [ ] Obsolete/unused files have been successfully deleted from the repository.
+
+## Out of Scope
+- Complete refactoring of `gui_2.py` into a fully modular system (this track only moves it, though preparing it for future non-monolithic structure is conceptually aligned).

conductor/archive/comprehensive_gui_ux_20260228/index.md

@@ -0,0 +1,5 @@
+# Track comprehensive_gui_ux_20260228 Context
+
+- [Specification](./spec.md)
+- [Implementation Plan](./plan.md)
+- [Metadata](./metadata.json)

conductor/archive/comprehensive_gui_ux_20260228/metadata.json

@@ -0,0 +1,10 @@
+{
+    "description":  "Enhance existing MMA orchestration GUI: tier stream panels, DAG editing, cost tracking, conductor lifecycle forms, track-scoped discussions, approval indicators, visual polish.",
+    "track_id":  "comprehensive_gui_ux_20260228",
+    "type":  "feature",
+    "created_at":  "2026-03-01T08:42:57Z",
+    "status":  "completed",
+    "updated_at":  "2026-03-01T20:15:00Z",
+    "refined_by":  "claude-opus-4-6 (1M context)",
+    "refined_from_commit":  "08e003a"
+}

conductor/archive/comprehensive_gui_ux_20260228/plan.md

@@ -0,0 +1,58 @@
+# Implementation Plan: Comprehensive Conductor & MMA GUI UX
+
+Architecture reference: [docs/guide_architecture.md](../../docs/guide_architecture.md), [docs/guide_mma.md](../../docs/guide_mma.md)
+
+## Phase 1: Tier Stream Panels & Approval Indicators
+Focus: Make all 4 tier output streams visible and indicate pending approvals.
+
+- [x] Task 1.1: Replace the single Tier 1 strategy text box in `_render_mma_dashboard` (gui_2.py:2700-2701) with four collapsible sections — one per tier. Each section uses `imgui.collapsing_header(f"Tier {N}: {label}")` wrapping a `begin_child` scrollable region (200px height). Tier 1 = "Strategy", Tier 2 = "Tech Lead", Tier 3 = "Workers", Tier 4 = "QA". Tier 3 should aggregate all `mma_streams` keys containing "Tier 3" with ticket ID sub-headers. Each section auto-scrolls to bottom when new content arrives (track previous scroll position, scroll only if user was at bottom).
+- [x] Task 1.2: Add approval state indicators to the MMA dashboard. After the "Status:" line in `_render_mma_dashboard` (gui_2.py:2672-2676), check `self._pending_mma_spawn`, `self._pending_mma_approval`, and `self._pending_ask_dialog`. When any is active, render a colored blinking badge: `imgui.text_colored(ImVec4(1,0.3,0.3,1), "APPROVAL PENDING")` using `sin(time.time()*5)` for alpha pulse. Also add a `imgui.same_line()` button "Go to Approval" that scrolls/focuses the relevant dialog.
+- [x] Task 1.3: Write unit tests verifying: (a) `mma_streams` with keys "Tier 1", "Tier 2 (Tech Lead)", "Tier 3: T-001", "Tier 4 (QA)" are all rendered (check by mocking `imgui.collapsing_header` calls); (b) approval indicators appear when `_pending_mma_spawn is not None`.
+- [x] Task 1.4: Conductor - User Manual Verification 'Phase 1: Tier Stream Panels & Approval Indicators' (Protocol in workflow.md)
+
+## Phase 2: Cost Tracking & Enhanced Token Table
+Focus: Add cost estimation to the existing token usage display.
+
+- [x] Task 2.1: Create a new module `cost_tracker.py` with a `MODEL_PRICING` dict mapping model name patterns to `{"input_per_mtok": float, "output_per_mtok": float}`. Include entries for: `gemini-2.5-flash-lite` ($0.075/$0.30), `gemini-2.5-flash` ($0.15/$0.60), `gemini-3-flash-preview` ($0.15/$0.60), `gemini-3.1-pro-preview` ($3.50/$10.50), `claude-*-sonnet` ($3/$15), `claude-*-opus` ($15/$75), `deepseek-v3` ($0.27/$1.10). Function: `estimate_cost(model: str, input_tokens: int, output_tokens: int) -> float` that does pattern matching on model name and returns dollar cost.
+- [x] Task 2.2: Extend the token usage table in `_render_mma_dashboard` (gui_2.py:2685-2699) from 3 columns to 5: add "Est. Cost" and "Model". Populate using `cost_tracker.estimate_cost()` with the model name from `self.mma_tier_usage` (need to extend `tier_usage` dict in `ConductorEngine._push_state` to include model name per tier, or use a default mapping: Tier 1 → `gemini-3.1-pro-preview`, Tier 2 → `gemini-3-flash-preview`, Tier 3 → `gemini-2.5-flash-lite`, Tier 4 → `gemini-2.5-flash-lite`). Show total cost row at bottom.
+- [x] Task 2.3: Write tests for `cost_tracker.estimate_cost()` covering all model patterns and edge cases (unknown model returns 0).
+- [x] Task 2.4: Conductor - User Manual Verification 'Phase 2: Cost Tracking & Enhanced Token Table' (Protocol in workflow.md)
+
+## Phase 3: Track Proposal Editing & Conductor Lifecycle Forms
+Focus: Make track proposals editable and add conductor setup/newTrack GUI forms.
+
+- [x] Task 3.1: Enhance `_render_track_proposal_modal` (gui_2.py:2146-2173) to make track titles and goals editable. Replace `imgui.text_colored` for title with `imgui.input_text(f"##track_title_{idx}", track['title'])`. Replace `imgui.text_wrapped` for goal with `imgui.input_text_multiline(f"##track_goal_{idx}", track['goal'], ImVec2(-1, 60))`. Add a "Remove" button per track (`imgui.button(f"Remove##{idx}")`) that pops from `self.proposed_tracks`. Edited values must be written back to `self.proposed_tracks[idx]`.
+- [x] Task 3.2: Add a "Conductor Setup" collapsible section at the top of the MMA dashboard (before the Track Browser). Contains a "Run Setup" button. On click, reads `conductor/workflow.md`, `conductor/tech-stack.md`, `conductor/product.md` using `Path.read_text()`, computes a readiness summary (files found, line counts, track count via `project_manager.get_all_tracks()`), and displays it in a read-only text region. This is informational only — no backend changes.
+- [x] Task 3.3: Add a "New Track" form below the Track Browser. Fields: track name (input_text), description (input_text_multiline), type dropdown (feature/chore/fix via `imgui.combo`). "Create" button calls a new helper `_cb_create_track(name, desc, type)` that: creates `conductor/tracks/{name}_{date}/` directory, writes a minimal `spec.md` from the description, writes an empty `plan.md` template, writes `metadata.json` with the track ID/type/status="new", then refreshes `self.tracks` via `project_manager.get_all_tracks()`.
+- [x] Task 3.4: Write tests for track creation helper: verify directory structure, file contents, and metadata.json format. Test proposal modal editing by verifying `proposed_tracks` list is mutated correctly.
+- [x] Task 3.5: Conductor - User Manual Verification 'Phase 3: Track Proposal Editing & Conductor Lifecycle Forms' (Protocol in workflow.md)
+
+## Phase 4: DAG Editing & Track-Scoped Discussion
+Focus: Allow GUI-based ticket manipulation and track-specific discussion history.
+
+- [x] Task 4.1: Add an "Add Ticket" button below the Task DAG section in `_render_mma_dashboard`. On click, show an inline form: ticket ID (input_text, default auto-increment like "T-NNN"), description (input_text_multiline), target_file (input_text), depends_on (multi-select or comma-separated input of existing ticket IDs). "Create" button appends a new `Ticket` dict to `self.active_tickets` with `status="todo"` and triggers `_push_mma_state_update()` to synchronize the ConductorEngine. Cancel hides the form. Store the form visibility in `self._show_add_ticket_form: bool`.
+- [x] Task 4.2: Add a "Delete" button to each DAG node in `_render_ticket_dag_node` (gui_2.py:2770-2773, after the Skip button). On click, show a confirmation popup. On confirm, remove the ticket from `self.active_tickets`, remove it from all other tickets' `depends_on` lists, and push state update. Only allow deletion of `todo` or `blocked` tickets (not `in_progress` or `completed`).
+- [x] Task 4.3: Add track-scoped discussion support. In `_render_discussion_panel` (gui_2.py:2295-2483), add a toggle checkbox "Track Discussion" (visible only when `self.active_track` is set). When toggled ON: load history via `project_manager.load_track_history(self.active_track.id, base_dir)` into `self.disc_entries`, set a flag `self._track_discussion_active = True`. When toggled OFF or track changes: restore project discussion. On save/flush, if `_track_discussion_active`, write to track history file instead of project history.
+- [x] Task 4.4: Write tests for: (a) adding a ticket updates `active_tickets` and has correct default fields; (b) deleting a ticket removes it from all `depends_on` references; (c) track discussion toggle switches `disc_entries` source.
+- [x] Task 4.5: Conductor - User Manual Verification 'Phase 4: DAG Editing & Track-Scoped Discussion' (Protocol in workflow.md)
+
+## Phase 5: Visual Polish & Integration Testing
+Focus: Dense, responsive dashboard with arcade aesthetics and end-to-end verification.
+
+- [x] Task 5.1: Add color-coded styling to the Track Browser table. Status column uses colored text: "new" = gray, "active" = yellow, "done" = green, "blocked" = red. Progress bar uses `imgui.push_style_color` to tint: <33% red, 33-66% yellow, >66% green.
+- [x] Task 5.2: Improve the DAG tree nodes with status-colored left borders. Use `imgui.get_cursor_screen_pos()` and `imgui.get_window_draw_list().add_rect_filled()` to draw a 4px colored strip to the left of each tree node matching its status color.
+- [x] Task 5.3: Add a "Dashboard Summary" header line at the top of `_render_mma_dashboard` showing: `Track: {name} | Tickets: {done}/{total} | Cost: ${total_cost:.4f} | Status: {mma_status}` in a single dense line with colored segments.
+- [x] Task 5.4: Write an end-to-end integration test (extending `tests/visual_sim_mma_v2.py` or creating `tests/visual_sim_gui_ux.py`) that verifies via `ApiHookClient`: (a) track creation form produces correct directory structure; (b) tier streams are populated during MMA execution; (c) approval indicators appear when expected; (d) cost tracking shows non-zero values after execution.
+- [x] Task 5.5: Verify all new UI elements maintain >30 FPS via `get_ui_performance` during a full MMA simulation run.
+- [x] Task 5.6: Conductor - User Manual Verification 'Phase 5: Visual Polish & Integration Testing' (Protocol in workflow.md)
+
+## Phase 6: Live Worker Streaming & Engine Enhancements
+Focus: Make MMA execution observable in real-time and configurable from the GUI. Currently workers are black boxes until completion.
+
+- [x] Task 6.1: Wire `ai_client.comms_log_callback` to per-ticket streams during `run_worker_lifecycle` (multi_agent_conductor.py:207-300). Before calling `ai_client.send()`, set `ai_client.comms_log_callback` to a closure that pushes intermediate text chunks to the GUI via `_queue_put(event_queue, loop, "response", {"text": chunk, "stream_id": f"Tier 3 (Worker): {ticket.id}", "status": "streaming..."})`. After `send()` returns, restore the original callback. This gives real-time output streaming to the Tier 3 stream panels from Phase 1.
+- [x] Task 6.2: Add per-tier model configuration to the MMA dashboard. Below the token usage table in `_render_mma_dashboard`, add a collapsible "Tier Model Config" section with 4 rows (Tier 1-4). Each row: tier label + `imgui.combo` dropdown populated from `ai_client.list_models()` (cached). Store selections in `self.mma_tier_models: dict[str, str]` with defaults from `mma_exec.get_model_for_role()`. On change, write to `self.project["mma"]["tier_models"]` for persistence.
+- [x] Task 6.3: Wire per-tier model config into the execution pipeline. In `ConductorEngine.run` (multi_agent_conductor.py:105-135), when creating `WorkerContext`, read the model name from the GUI's `mma_tier_models` dict (passed via the event queue or stored on the engine). Pass it through to `run_worker_lifecycle` which should use it in `ai_client.set_provider`/`ai_client.set_model_params` before calling `send()`. Also update `mma_exec.py:get_model_for_role` to accept an override parameter.
+- [x] Task 6.4: Add parallel DAG execution. In `ConductorEngine.run` (multi_agent_conductor.py:100-135), replace the sequential `for ticket in ready_tasks` loop with `asyncio.gather(*[loop.run_in_executor(None, run_worker_lifecycle, ...) for ticket in ready_tasks])`. Each worker already gets its own `ai_client.reset_session()` so they're isolated. Guard with `ai_client._send_lock` awareness — if the lock serializes all sends, parallel execution won't help. In that case, create per-worker provider instances or use separate session IDs. Mark this task as exploratory — if `_send_lock` blocks parallelism, document the constraint and defer.
+- [x] Task 6.5: Add automatic retry with model escalation. In `ConductorEngine.run`, after `run_worker_lifecycle` returns, check if `ticket.status == "blocked"`. If so, and `retry_count < max_retries` (default 2), increment retry count, escalate the model (e.g., flash-lite → flash → pro), and re-execute. Store `retry_count` as a field on the ticket dict. After max retries, leave as blocked.
+- [x] Task 6.6: Write tests for: (a) streaming callback pushes intermediate content to event queue; (b) per-tier model config persists to project TOML; (c) retry escalation increments model tier.
+- [x] Task 6.7: Conductor - User Manual Verification 'Phase 6: Live Worker Streaming & Engine Enhancements' (Protocol in workflow.md)