What is the benefit of multi-model deliberation?

Multi-model deliberation reduces individual model bias and hallucination by requiring multiple AI perspectives to reach a consensus, ensuring higher reliability for code verification.

How does the skill handle versioning and reproducibility?

The skill utilizes snapshot pinning via Git commit SHAs to ensure that every verification result is tied to a specific state of the codebase for a clear audit trail.

Does it support specific security audits?

Yes, you can use the 'rubric_focus' parameter to specifically target Security, Performance, or Accessibility during the verification process.

Can I integrate this skill into my CI/CD pipeline?

Yes, the skill provides standardized exit codes (0 for PASS, 1 for FAIL, 2 for UNCLEAR) specifically designed for integration into automated build and deployment workflows.

LLM Council Verification

Name: LLM Council Verification
Author: amiable-dev

byamiable-dev

•

セキュリティとテスト

Verifies code and documentation using multi-model deliberation to ensure high-confidence consensus on implementation quality.

The LLM Council Verification skill enhances Claude Code with a sophisticated multi-model deliberation system for validating technical work. By moving beyond single-model assessments, it facilitates a 'council' of models to reach consensus on the correctness, security, and completeness of code changes or documentation. This skill is particularly useful for high-stakes environments where objective validation is required, providing machine-actionable JSON verdicts, confidence scores, and detailed audit trails that can be integrated directly into CI/CD pipelines or manual review workflows.

主な機能

01Multi-model consensus for objective validation

02Targeted rubrics for Security, Performance, and Accessibility

03Machine-actionable exit codes for CI/CD integration

041 GitHub stars

05Structured JSON verdicts with confidence scoring

06Snapshot-based reproducibility via Git commit SHAs

ユースケース

01Automated quality gating in production deployment pipelines

02Validating security-critical code changes before commit

03Auditing technical documentation for completeness and accuracy

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add amiable-dev/luminescent-cluster council-verify

For use in Claude.ai and ChatGPT

Download Skill