AI Roundtable

A Multi-AI debate platform that enables structured discussions between different AI models (Claude, ChatGPT, Gemini, Perplexity) through the Model Context Protocol (MCP).

Key Features

• 4 AI Providers: Claude (Anthropic), ChatGPT (OpenAI), Gemini (Google), Perplexity
• 7 Debate Modes: Collaborative, Adversarial, Socratic, Expert Panel, Devil's Advocate, Delphi, Red Team/Blue Team
• AI-Powered Analysis: Semantic consensus analysis using lightweight AI models
• Native Web Search: Each agent uses its provider's native search capability
• MCP Protocol: Standard interface for Claude Desktop and other MCP clients

Quick Start

Installation

# Option 1: Run directly with npx (Recommended)
npx github:mnthe/ai-roundtable

# Option 2: Clone and build locally
git clone https://github.com/mnthe/ai-roundtable.git
cd ai-roundtable
pnpm install
pnpm build

Environment Setup

Create a .env file with your API keys:

# Required: At least one API key
ANTHROPIC_API_KEY=sk-ant-...     # For Claude agents
OPENAI_API_KEY=sk-...            # For ChatGPT agents
GOOGLE_API_KEY=...               # For Gemini agents
PERPLEXITY_API_KEY=pplx-...      # For Perplexity agents

MCP Client Configuration

Add to your MCP client configuration (e.g., Claude Desktop, IDE plugins):

{
  "mcpServers": {
    "ai-roundtable": {
      "command": "npx",
      "args": ["-y", "github:mnthe/ai-roundtable"],
      "env": {
        "ANTHROPIC_API_KEY": "sk-ant-...",
        "OPENAI_API_KEY": "sk-...",
        "GOOGLE_API_KEY": "...",
        "PERPLEXITY_API_KEY": "pplx-..."
      }
    }
  }
}

MCP Tools

Tool	Description
start_roundtable	Start a new debate session
continue_roundtable	Continue existing debate with additional rounds
get_consensus	Analyze consensus level in a session
get_agents	List available AI agents
list_sessions	List debate sessions with filters
get_round_details	Get responses for a specific round
get_response_detail	Get detailed response from a specific agent
get_citations	Get citations from the debate
get_thoughts	Get agent's reasoning evolution
synthesize_debate	AI-powered debate synthesis
export_session	Export session (markdown/JSON)
control_session	Pause/resume/stop session

See API Reference for detailed parameters and responses.

Debate Modes

Mode	Execution	Use Case
Collaborative	Parallel	Consensus building, brainstorming
Adversarial	Sequential	Stress-testing ideas
Socratic	Sequential	Deep exploration via questions
Expert Panel	Parallel	Independent expert assessments (supports custom perspectives)
Devil's Advocate	Sequential	Preventing groupthink
Delphi	Parallel	Anonymized iterative consensus
Red Team/Blue Team	Hybrid	Security/risk analysis

See Architecture for detailed mode descriptions and execution patterns.

Documentation

Document	Description
Architecture	System architecture, debate flow, 4-layer response structure
API Reference	Complete API documentation, types, error handling
Development Guide	Contributing, testing, adding providers/modes
Testing Guide	Testing patterns and practices

Development

pnpm install      # Install dependencies
pnpm build        # Build
pnpm dev          # Development (watch mode)
pnpm test         # Run tests
pnpm typecheck    # Type check
pnpm lint         # Lint

License

MIT

Contributing

1. Fork the repository
2. Create a feature branch: git checkout -b feature/my-feature
3. Make changes with tests
4. Run checks: pnpm typecheck && pnpm lint && pnpm test
5. Push and create PR

See Development Guide for detailed contribution guidelines.