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

docs(mma): Draft Multi-Model Delegation and finish Proposal

Browse Source
This commit is contained in:
Edward R. Gonzalez
2026-02-24 22:35:02 -05:00
parent baff5c18d3
commit f62bf3113f
1 changed files with 11 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
conductor/tracks/mma_implementation_20260224/proposal.md
View File

@@ -1,7 +1,7 @@
# Conductor Self-Reflection & Upgrade Strategy Proposal # Conductor Self-Reflection & Upgrade Strategy Proposal
## 1. Executive Summary ## 1. Executive Summary
*To be drafted.* This proposal outlines a strategic path for upgrading the Gemini CLI `conductor` extension to fully embrace the 4-Tier Hierarchical Multi-Model Architecture principles. By migrating from a monolithic, context-heavy single-agent loop to a compartmentalized, multi-model delegation system, Conductor can drastically reduce token burn, mitigate hallucination loops, and grant developers surgical Human-In-The-Loop (HITL) control over execution tasks.
## 2. Memory Siloing & Token Firewalling ## 2. Memory Siloing & Token Firewalling
@@ -26,7 +26,15 @@ To enforce precise developer control, Conductor should natively embed a **Human-
3. **Rollback Mutators:** Provide quick access commands (e.g., via `after_tool_call`) to reject the change, auto-restoring the last known file state, and feeding the error/feedback directly back to the model without breaking the run loop. 3. **Rollback Mutators:** Provide quick access commands (e.g., via `after_tool_call`) to reject the change, auto-restoring the last known file state, and feeding the error/feedback directly back to the model without breaking the run loop.
## 4. Multi-Model/Sub-Agent Delegation ## 4. Multi-Model/Sub-Agent Delegation
*To be drafted. Focus: Evaluating reliance on single models and proposing a design for delegating tasks via Gemini CLI Sub-agents and Skills.*
### Current Evaluation
Conductor heavily relies on the single primary LLM instantiated by the Gemini CLI session. When acting as a PM, Tech Lead, and Worker simultaneously, the model experiences extreme context exhaustion. Furthermore, handling minor formatting, syntax repairs, or summaries with expensive high-tier reasoning models results in suboptimal cost-efficiency.
### Proposed Upgrade Strategy
Conductor should leverage the native **Sub-Agent & Skill Routing capabilities**:
1. **Dynamic Tier Routing:** Utilize specific Sub-agents (like `codebase_investigator` for planning/AST generation) and custom Skills for discrete tasks.
2. **Stateless Utility Agents (Tier 4):** Hook into test runner commands via `after_tool_call`. If `pytest` fails with massive `stderr`, immediately invoke a cheap background utility sub-agent to parse the log and return a condensed 20-word summary back to the main Orchestrator, rather than feeding the main Orchestrator raw traceback tokens.
3. **Contract Stubbers:** Embed `contract_stubber` skills that explicitly limit a sub-agent's action strictly to writing `class` or `def` definitions, ensuring cross-module dependency generation without full implementation drift.
## 5. Implementation Strategy ## 5. Implementation Strategy
*To be drafted. Focus: Alignment with 4-Tier Architecture philosophy and leveraging Gemini CLI's extension framework (manifests, hooks, sub-agents).* These upgrades can be realized by augmenting the `gemini-extension.json` manifest with designated MCP hooks, adding new custom Skills to `~/.gemini/skills/`, and overriding default CLI execution flows with `before_tool_call` and `after_tool_call` interception logic tailored explicitly for Token Firewalling and Execution Checkpoints.