The Council orchestrates four specialized Claude agents through the Model Context Protocol, routing tasks automatically based on complexity signals. Simple formatting tasks go to the lightweight Aide, straightforward work to the Executor, strategic planning to the Chancellor, and all outputs are reviewed by the Supervisor. Routing uses keyword and word-count heuristics without extra LLM calls. Supports persistent sessions via memory, file, or SQLite storage.
Cognium trust score
50%
Tier
Unverified
Composite of vulnerability cleanliness, spec conformance, provenance, stability, and usage signals — scanned and weighted by Cognium. Human and agent signals are tracked separately.
Returns 7 tools: search_skills, get_skill, list_leaderboard, get_trust_breakdown, resolve_composition, plus the ChatGPT-connector search and fetch. Every tool is annotated read-only.
Resolve this skill directly via MCP tools/call get_skill.