What kind of issues can this skill detect?

Beyond syntax, it detects logic flaws, security vulnerabilities, performance bottlenecks, and inconsistencies with project requirements using specialized rubrics.

Can I use Council Verify in my CI/CD pipeline?

Yes, the skill is designed for automation and returns specific exit codes (0 for PASS, 1 for FAIL, 2 for UNCLEAR) that can trigger or block deployment workflows.

Does it keep a record of the verification process?

Yes, it creates a persistent audit trail and stores transcripts, allowing teams to review the rationale behind every pass or fail verdict.

How does confidence thresholding work?

You can set a minimum confidence score (e.g., 0.7). If the multi-model consensus falls below this number, the verdict is marked as 'unclear' and flagged for human review.

What is multi-model deliberation in Council Verify?

Multi-model deliberation involves multiple AI models reviewing the same code or document to reach a consensus, reducing hallucinations and increasing the accuracy of the final verdict.

Council Verification

Name: Council Verification
Author: amiable-dev

byamiable-dev

•

セキュリティとテスト

Verifies code, documents, and implementations using multi-model LLM deliberation to provide objective, high-confidence quality verdicts.

概要

Council Verification leverages the LLM Council framework to perform sophisticated multi-model reviews of your codebase and documentation. By utilizing consensus-based deliberation, the skill generates structured, machine-actionable verdicts that go beyond simple linting to assess logic, security, and requirement compliance. It is an essential tool for developers who need high-assurance validation before committing changes, offering reproducible audit trails and specific rubrics for security, performance, and accessibility.

主な機能

Snapshot pinning using Git commit SHAs for reproducibility
CI/CD-ready exit codes for automated quality gates
Multi-model consensus deliberation for high-confidence results
Structured JSON output with confidence scores and rubric ratings
Targeted verification focus for Security, Performance, and Accessibility
1 GitHub stars

ユースケース

Establishing a persistent audit trail for AI-assisted code governance
Validating complex implementations against original requirement documents
Automating pre-commit code reviews for security and logic errors

概要

主な機能

Snapshot pinning using Git commit SHAs for reproducibility
CI/CD-ready exit codes for automated quality gates
Multi-model consensus deliberation for high-confidence results
Structured JSON output with confidence scores and rubric ratings
Targeted verification focus for Security, Performance, and Accessibility
1 GitHub stars

ユースケース

Establishing a persistent audit trail for AI-assisted code governance
Validating complex implementations against original requirement documents
Automating pre-commit code reviews for security and logic errors