spyrae/claude-concilium

Overview

A multi-agent consultation framework that integrates with Claude Code via the Model Context Protocol (MCP). It routes queries to external AI models like OpenAI, Gemini, Qwen, and DeepSeek to provide second opinions within Claude's environment.

Best for

Best for
Developers using Claude Code who want to verify or enrich outputs with other AI models

Use cases

• Cross-checking Claude's output with another model for accuracy
• Comparing reasoning from multiple AI models on a single prompt
• Building multi-model validation into a Claude Code workflow

How to use

Tools exposed

• codex
• gemini
• qwen

Tested with

Claude Code

Example client config

{\n  "mcpServers": {\n    "mcp-openai": {\n      "type": "stdio",\n      "command": "node",\n      "args": ["/absolute/path/to/servers/mcp-openai/server.js"],\n      "env": {\n        "CODEX_HOME": "~/.codex-minimal"\n      }\n    },\n    "mcp-gemini": {\n      "type": "stdio",\n      "command": "node",\n      "args": ["/absolute/path/to/servers/mcp-gemini/server.js"]\n    }\n  }\n}

Notes

A multi-agent consultation framework that integrates with Claude Code via the Model Context Protocol (MCP). It routes queries to external AI models like OpenAI, Gemini, Qwen, and DeepSeek to provide second opinions within Claude’s environment.

8 stars on GitHub. Last updated 2026-03-02. Licensed MIT.

Use cases

• Cross-checking Claude’s output with another model for accuracy
• Comparing reasoning from multiple AI models on a single prompt
• Building multi-model validation into a Claude Code workflow

Pros

• Leverages MCP for seamless integration with Claude Code
• Supports multiple major AI providers out of the box
• Lightweight JavaScript implementation with low overhead

Cons

• Requires API keys for each external model used
• Adds latency from sequential model calls
• Limited to models with compatible API endpoints

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.