amiable-devamiable-devCommunityยท3 Agent Skills Included

llm-council

Multi-model code review, verification, and CI quality gates

Runs code reviews, requirement verification, and CI/CD quality gates through multi-model deliberation with anonymized peer ranking. Eliminates single-model blind spots and biased self-review by having several models score changes against structured rubrics. Returns machine-readable pass, fail, or unclear verdicts with blocking issues, confidence scores, and full audit transcripts for pipelines.
npx skills add amiable-dev/llm-council --all -g -y
Available:

Gives the agent a durable architecture map of the council system, explaining how the tier, triage, deliberation, and gateway layers work together so it can operate and extend the verification skills correctly.

All Skills in This Repository (3)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install LLM Council?โ–ผ

Run `npx skills add amiable-dev/llm-council --all -g -y` in your terminal to install all skills in this suite globally.

How does LLM Council verify code changes?โ–ผ

It sends your code snapshot to multiple models in parallel, has them anonymously peer-rank each other's assessments, then a chairman model synthesizes a final pass, fail, or unclear verdict with structured scores.

Can I use LLM Council as a CI/CD quality gate?โ–ผ

Yes. The council-gate skill returns exit codes 0 for pass, 1 for fail, and 2 for unclear, and includes a ready-made GitHub Actions workflow example.

What API keys does LLM Council need?โ–ผ

It works with a single OpenRouter key for 100+ models, or direct provider keys for Anthropic, OpenAI, and Google, stored securely in your system keychain or environment variables.

Does LLM Council work with Claude Code and other MCP clients?โ–ผ

Yes. It ships as an MCP server with bundled skills that run in Claude Code and any MCP-compatible client, plus CLI and HTTP API options.

Related Repositories in Software Engineering

View All in Software Engineeringโ†’