Councly

Runs MCP hearings with multiple LLMs to debate topics and synthesize verdicts for governance and code reviews.

javascript

0

GitHub Stars

javascript

Language

7 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "slmnsrf-councly-mcp": {
      "command": "npx",
      "args": [
        "@councly/mcp"
      ],
      "env": {
        "COUNCLY_API_KEY": "cnc_your_key_here",
        "COUNCLY_BASE_URL": "https://councly.ai"
      }
    }
  }
}

You run an MCP server that lets multiple LLMs engage in council hearings, debate topics, and synthesize verdicts. This server bridges Claude Code, Codex, and other MCP-compatible assistants, enabling collaborative reasoning and structured results through a simple, scalable setup.

How to use

Install and start the MCP server, then connect your MCP client to initiate hearings. You can create hearings on any topic within the allowed length and choose a workflow that fits your needs, such as auto-discussion, focused review, or brainstorming. The server returns a clear verdict, a trust score, and perspectives from each counsel, along with the total cost in credits.

How to install

Prerequisites: you need Node.js installed along with npm. Ensure you have network access to install packages.

# Install the MCP server globally
npm install -g @councly/mcp

# Or run directly without installing
npx @councly/mcp

Configuration and usage notes

Set up your API key and configure your MCP client to connect to the server. The key authorizes requests to create hearings and retrieve results. You can provide the key via environment variables or client configuration.

Environment variables to consider include COUNCLY_API_KEY for authentication and COUNCLY_BASE_URL to override the default API endpoint if needed.

Examples and tips

Create a hearing to review a code snippet or discuss a topic in depth. Adjust the preset and workflow to balance speed and thoroughness. Use the status tool to check on ongoing hearings and verify completion details.

Security and usage notes

Keep your API key secure. Do not share it in public code or logs. Monitor hearing activity and review verdicts to ensure they align with your governance requirements.

Available tools

council_hearing

Create a council hearing where multiple LLMs debate a topic and produce a synthesized verdict with counsel perspectives and a trust score.

council_status

Query the status of an existing hearing by its hearing_id to obtain progress, verdict, and cost information.