A
licenseA
qualityA
maintenanceAn MCP server that seats multiple LLMs as a council, letting your assistant query them in parallel or in sequence, relay answers for cross-critique, and merge conclusions within one conversation.
6
38 PyPI
2
MIT
llm-council
by blessthis
Related Servers
Alternatives to llm-council
No user-submitted related servers found.
Related Servers
- AlicenseNot gradedqualityCmaintenanceMulti-model AI council MCP server that lets multiple LLMs debate, critique, and vote on questions to reach a consensus, with automatic failover and hot-swap of models during debates.MIT
- AlicenseNot gradedqualityCmaintenanceAn MCP server that enables an AI orchestrator to convene a council of other AIs via their CLIs, deliberate, and synthesize consensus.1 npmGPL 3.0
- AlicenseNot gradedqualityDmaintenanceMCP server for AI agents to conduct multi-LLM roundtable discussions, returning structured common, divergent, and unique perspectives.MIT
- AlicenseAqualityBmaintenanceMCP server that reduces confirmation bias in LLMs by orchestrating structured debates between asymmetric context sessions.13MIT
- AlicenseAqualityDmaintenanceMCP server orchestrating local CLI agents (Claude Code, OpenAI Codex, Google Gemini) for cross-validation, second opinions, and persona-driven prompting.18MIT
TDQS
A4/5.0
Scored across 17 tools
Disambiguation5/5
The tools are clearly separated into council and chat workflows, plus supporting utilities. Within each group, actions like poll, answer, ask, and score are unambiguous. The only near-overlap (council_poll vs. council_is_model_replied) is resolved by distinct purposes—status reporting vs. model-presence check.
Naming Consistency5/5
All tools follow a consistent snake_case convention with clear prefixes: council_* for council operations, chat_* for chat operations, and descriptive nouns like model_scores, seat_health, list_seats. The pattern is predictable and uniform across the surface.
Tool Count4/5
With 17 tools, the count is slightly above the typical well-scoped range, but the domain spans two distinct workflows (councils and chats) plus seat/quality management, so each tool serves a clear purpose without redundancy. The bulk is justified by the multi-step nature of both workflows.
Completeness5/5
The council lifecycle is fully covered from start to poll, answer, cross-examine, score, reveal, and close, including blind checks and persistence. The chat workflow also has full CRUD and history. Seat management and model quality feedback round out the surface with no obvious gaps.
Maintenance
ActivityMaintained
ResponsivenessNo issues