
from Claude Code Skills11
Orchestrate a diverse council of external AI models (Gemini, Codex, Qwen, GLM, Kimi) for consensus-driven code reviews and architectural decisions.
This skill transforms a single AI agent into an orchestrator for a multi-model 'council'. It allows the agent to solicit perspectives from various external LLMs to reduce hallucination, identify blind spots, and reach a consensus on critical technical decisions.
Use this skill when you need a high-confidence review of a complex PR, an architectural critique of a new system design, or an adversarial analysis of a security-critical piece of code. It is explicitly triggered by commands like /council or requests to 'consult the council'.
Designed primarily for Claude Code and other agents capable of spawning sub-agents and executing CLI tools.
This skill has not been reviewed by our automated audit pipeline yet.