ed/manual_slop
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

chore(conductor): Fix SKILL.md and documentation typos to correctly use the new Role-Based sub-agent protocol

Browse Source
This commit is contained in:
Edward R. Gonzalez
2026-02-25 09:15:25 -05:00
parent e60eef5df8
commit 3f9c6fc6aa
2 changed files with 109 additions and 109 deletions
Download Patch File Download Diff File Expand all files Collapse all files

90
MMA_Support/Architecture_Recommendation.md
View File

@@ -1,45 +1,45 @@
# MMA Hierarchical Delegation: Recommended Architecture # MMA Hierarchical Delegation: Recommended Architecture
## 1. Overview ## 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. 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 ## 2. Agent Roles & Responsibilities
### Tier 2: The Conductor (Tech Lead) ### Tier 2: The Conductor (Tech Lead)
- **Role:** Orchestrator of the project lifecycle via the Conductor framework. - **Role:** Orchestrator of the project lifecycle via the Conductor framework.
- **Context:** High-reasoning, long-term memory of project goals and specifications. - **Context:** High-reasoning, long-term memory of project goals and specifications.
- **Key Tool:** `mma-orchestrator` skill (Strategy). - **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. - **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) ### Tier 3: The Worker (Contributor)
- **Role:** Stateless code generator. - **Role:** Stateless code generator.
- **Context:** Isolated. Sees only the target file and the specific ticket. - **Context:** Isolated. Sees only the target file and the specific ticket.
- **Protocol:** Receives a "Worker" system prompt. Outputs clean code or diffs. - **Protocol:** Receives a "Worker" system prompt. Outputs clean code or diffs.
- **Invocation:** `.\scripts - **Invocation:** `.\scripts\run_subagent.ps1 -Role Worker -Prompt "..."`
un_subagent.ps1 -Role Worker -Prompt "..."`
### Tier 4: The Utility (QA/Compressor)
### Tier 4: The Utility (QA/Compressor) - **Role:** Stateless translator and summarizer.
- **Role:** Stateless translator and summarizer. - **Context:** Minimal. Sees only the error trace or snippet.
- **Context:** Minimal. Sees only the error trace or snippet. - **Protocol:** Receives a "QA" system prompt. Outputs compressed findings (max 50 tokens).
- **Protocol:** Receives a "QA" system prompt. Outputs compressed findings (max 50 tokens). - **Invocation:** `.\scripts\run_subagent.ps1 -Role QA -Prompt "..."`
- **Invocation:** `.\scripts
un_subagent.ps1 -Role QA -Prompt "..."` ## 3. Invocation Protocol
## 3. Invocation Protocol ### Step 1: Detection
Tier 2 detects a delegation trigger:
### Step 1: Detection - Coding task > 50 lines.
Tier 2 detects a delegation trigger: - Error trace > 100 lines.
- Coding task > 50 lines.
- Error trace > 100 lines. ### Step 2: Spawning
Tier 2 calls the delegation script:
### Step 2: Spawning ```powershell
Tier 2 calls the delegation script: .\scripts\run_subagent.ps1 -Role <Worker|QA> -Prompt "Specific instructions..."
```powershell ```
.\scripts
un_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.
### Step 3: Integration - **If QA:** Tier 2 uses the compressed error to inform the next fix attempt or passes it to a Worker.
Tier 2 receives the sub-agent's response.
- **If Worker:** Tier 2 applies the code changes (using `replace` or `write_file`) and verifies. ## 4. System Prompt Management
- **If QA:** Tier 2 uses the compressed error to inform the next fix attempt or passes it to a Worker. 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).

128
mma-orchestrator/SKILL.md
View File

@@ -1,64 +1,64 @@
--- ---
name: mma-orchestrator name: mma-orchestrator
description: Enforces the 4-Tier Hierarchical Multi-Model Architecture (MMA) within Gemini CLI using Token Firewalling and sub-agent task delegation. 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 # 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). 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`. 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:** **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: 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 `.\scripts\run_subagent.ps1 -Role <Role> -Prompt "..."`
un_subagent.ps1 -Prompt "..."`
## 1. The Tier 3 Worker (Heads-Down Coding)
## 1. The Tier 3 Worker (Heads-Down Coding) When you need to perform a significant code modification (e.g., refactoring a 50-line+ script, writing a massive class, or implementing a predefined spec):
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.
1. **DO NOT** attempt to write or use `replace`/`write_file` yourself. Your history will bloat. 2. **DO** construct a single, highly specific prompt.
2. **DO** construct a single, highly specific prompt. 3. **DO** spawn a sub-agent using `run_shell_command` pointing to the target file.
3. **DO** spawn a sub-agent using `run_shell_command` pointing to the target file. *Command:* `.\scripts\run_subagent.ps1 -Role Worker -Prompt "Read [FILE_PATH] and modify it to implement [SPECIFIC_INSTRUCTION]. Only write the code, no pleasantries."`
*Command:* `.\scripts 4. The Tier 3 Worker is stateless and has no tool access. You must take the clean code it returns and apply it to the file system using your own `replace` or `write_file` tools.
un_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., 100+ lines of `stderr`):
## 2. The Tier 4 QA Agent (Error Translation) 1. **DO NOT** analyze the raw `stderr` in your own context window.
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`): 2. **DO** immediately spawn a stateless Tier 4 agent to compress the error.
1. **DO NOT** analyze the raw `stderr` in your own context window. 3. *Command:* `.\scripts\run_subagent.ps1 -Role QA -Prompt "Summarize this stack trace into a 20-word fix: [PASTE_SNIPPET_OF_STDERR_HERE]"`
2. **DO** immediately spawn a stateless Tier 4 agent to compress the error. 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. *Command:* `.\scripts
un_subagent.ps1 -Prompt "Summarize this stack trace into a 20-word fix: [PASTE_SNIPPET_OF_STDERR_HERE]"` ## 3. Context Amnesia (Phase Checkpoints)
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. When you complete a major Phase or Track within the `conductor` workflow:
1. Stage your changes and commit them.
## 3. Context Amnesia (Phase Checkpoints) 2. Draft a comprehensive summary of the state changes in a Git Note attached to the commit.
When you complete a major Phase or Track within the `conductor` workflow: 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.
1. Stage your changes and commit them.
2. Draft a comprehensive summary of the state changes in a Git Note attached to the commit. <examples>
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. ### Example 1: Spawning a Tier 4 QA Agent
**User / System:** `pytest tests/test_gui.py` failed with 400 lines of output.
<examples> **Agent (You):**
### Example 1: Spawning a Tier 4 QA Agent ```json
**User / System:** `pytest tests/test_gui.py` failed with 400 lines of output. {
**Agent (You):** "command": ".\\scripts\\run_subagent.ps1 -Role QA -Prompt \"Summarize this stack trace into a 20-word fix: [snip first 30 lines...]\"",
```json "description": "Spawning Tier 4 QA to compress error trace statelessly."
{ }
"command": ".\scripts ```
un_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
### Example 2: Spawning a Tier 3 Worker {
**User:** Please implement the `ASTParser` class in `file_cache.py` as defined in Track 1. "command": ".\\scripts\\run_subagent.ps1 -Role Worker -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.\"",
**Agent (You):** "description": "Delegating implementation to a Tier 3 Worker."
```json }
{ ```
"command": ".\scripts </examples>
un_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." <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).
</examples> - When explicitly instructed to act as a "Tech Lead" or "Orchestrator".
- When managing complex, multi-file Track implementations.
<triggers> </triggers>