autonomous-fix-review
Run autonomous multi-model code review on a finding. Invokes claude/codex/gemini via reviewd, evaluates consensus, and auto-commits or escalates to human.
- License 许可
- MIT
- 最近更新
- 2026-08-03
- Stars
- ★ 1
- Forks
- 0
01 前言 Front Matter
- name
- autonomous-fix-review
- context
- fork
- description
- Run autonomous multi-model code review on a finding. Invokes claude/codex/gemini via reviewd, evaluates consensus, and auto-commits or escalates to human.
- argument-hint
- <finding-file-or-description>
- user-invocable
- true
02 资源文件 2
- 参考文档SOURCE.md143 B
- 参考文档TRUST.auto.yaml2.2 KB
03 SKILL.md 正文
Autonomous Fix-Review Pipeline
Give it a finding, walk away, come back to a committed fix that passed multi-model review.
Usage
/autonomous-fix-review "Fix the null pointer in src/main.py:42"
/autonomous-fix-review docs/research/trail/findings/finding-001.yaml
7-Phase Pipeline
| Phase | Action | Tool |
|---|---|---|
| 1. Research | Gather context (docs, code) | /find-docs, /context7-mcp |
| 2. Implement | Dispatch coder in worktree with TDD | subagent-driven-development |
| 3. Review | Invoke claude + codex + gemini | reviewd.invoke_cli() |
| 4. Consensus | Evaluate 75% threshold | mde.consensus.evaluate_consensus() |
| 5. Decision | Apply autonomy mode | ConsensusGate.apply_autonomy() |
| 6. Quality gate | Run lint + test + validate | uv run mde-py quality |
| 7. Commit | Commit on feature branch or escalate | git commit |
Autonomy Modes
- supervised: Human approves everything except unanimous approval
- semi-autonomous (default): Auto-retry on warnings, escalate on disagreement
- autonomous: Proceed on any non-abort consensus
Decision Matrix
| Consensus | Supervised | Semi-Autonomous | Autonomous |
|---|---|---|---|
| >= 90% approve, no criticals | PROCEED | PROCEED | PROCEED |
| >= 75% approve, no criticals | ESCALATE | RETRY | PROCEED |
| Any critical findings | ESCALATE | ESCALATE | ESCALATE |
| < 75% approve | ESCALATE | ESCALATE | ESCALATE |
| All reject | ABORT | ABORT | ABORT |
Workflow
# Step 1: Parse the finding
finding=$(echo "$ARGUMENTS")
# Step 2: Research context
# Invoke /find-docs and /context7-mcp for the affected code area
# Step 3: Implement fix (if not already done)
# Dispatch coder subagent in worktree with TDD
# Step 4-5: Run multi-model review + consensus
uv run mde-py review "$finding" --autonomy semi-autonomous
# Step 6: Quality gate (auto-runs unless --skip-quality)
uv run mde-py quality
# Step 7: Based on decision output:
# PROCEED → commit on feature branch
# ESCALATE → present findings to human for decision
# RETRY → re-implement and re-review (up to max_retries)
# ABORT → stop, finding may be invalid
Constraints
- Zero API keys — all LLM calls via subscription CLIs (claude, codex, gemini)
- Worktree isolation for implementation phase
- Human interview ONLY on consensus disagreement (ESCALATE)
- Quality gate (6/6) required before commit
04 同仓库更多 Skills
next-workflow-step
Use this skill when the user wants to transition to the next workflow step, wrap up a session cleanly, prepare a handoff, or says things like "next step", "continue the workflow", "wrap up", "hand off", "move to the next phase", "what's next in embrace", or "prepare for /clear". Also triggers when the current octo:embrace phase appears complete and no explicit next action is set.
source-discovery
This skill should be used when the user asks to "find sources", "discover URLs", "catalog a source", "classify a URL", "search for libraries", "check what exists", or mentions source discovery, URL cataloging, source classification, or research source management.
review-synthesis
This skill should be used when the user asks to "summarize review findings", "prioritize fixes", "rank review issues", "combine reviewer feedback", "create action items from review", or mentions review synthesis, finding prioritization, severity ranking, or review result summarization.
research-scoring
This skill should be used when the user asks to "score research", "check research quality", "evaluate coverage", "measure improvement", "assess research completeness", or mentions research scoring, quality metrics, coverage gaps, improvement baseline, or research pipeline status.
notebooklm-integration
This skill should be used when the user asks to "ingest into NotebookLM", "query NotebookLM", "add sources to notebook", "cross-source synthesis", "check stale sources", or mentions NotebookLM, notebook integration, source ingestion, or knowledge consolidation.
evidence-synthesis
This skill should be used when the user asks to "synthesize findings", "combine research", "cross-reference sources", "summarize evidence", "write a deep review", or mentions evidence synthesis, finding consolidation, cross-source analysis, or research summarization.