amiable-dev/LLM Council Quality Gate
Multi-model consensus verification for AI-generated code and content
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Activelast commit Jan 30, 2026
- License
- MIT
Pinned Snippet
uses: amiable-dev/llm-council-action@e80e67589404cc2aaadfc2c06dd613b93206862e # v1.0.0tags can be moved; commit SHAs can't. why a SHA?
Inputs
| name | description | required | default |
|---|---|---|---|
| command | Command to run: gate, verify, or review | no | gate |
| snapshot | Git snapshot/SHA to verify | no | ${{ github.sha }} |
| file-paths | Specific file paths to verify (space-separated) | no | — |
| confidence-threshold | Minimum confidence score to pass (0.0-1.0) | no | 0.8 |
| rubric-focus | Focus area for evaluation: Security, Performance, Testing, or General | no | General |
| version | llm-council-core version to install (default: latest stable) | no | 0.24.21 |
| python-version | Python version to use | no | 3.11 |
| fail-on-unclear | Fail the action when verdict is UNCLEAR (exit code 2) | no | false |
Outputs
| name | description |
|---|---|
| verdict | Verdict: PASS, FAIL, or UNCLEAR |
| exit-code | Exit code: 0=PASS, 1=FAIL, 2=UNCLEAR |
| confidence | Aggregate confidence score (0.0-1.0) |
| summary | Human-readable summary of evaluation |
| transcript-path | Path to verification transcript (if generated) |