MCP Catalogs
← Home

roundtable

by askbudi·★ 107·Score 46

Zero-configuration MCP server coordinating multiple AI coding assistants through intelligent auto-discovery.

developer-toolsai-llmproductivity
15
Forks
8
Open issues
7 mo ago
Last commit
2d ago
Indexed

Overview

Roundtable AI is a local MCP server that allows users to delegate tasks from a single prompt in their IDE to specialized AI sub-agents like Gemini, Claude Code, Cursor, and Codex. The system maintains context continuity across all sub-agents, enables parallel execution of tasks, and leverages each model's strengths for specialized problem-solving. It uses existing CLI tools and API subscriptions without any markup, making it a cost-effective solution for complex engineering workflows.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

you:Debugging complex production issues by simultaneously analyzing logs from multiple AI agents
you:Performance optimization by leveraging different models for SQL analysis, frontend profiling, and code refactoring
you:Code review and implementation using specialized models for different programming languages and frameworks
you:Does Roundtable AI require separate API keys for each AI model?
you:How does Roundtable handle context continuity between different AI models?

When to choose this

When you need to leverage multiple specialized AI models simultaneously for complex engineering tasks without manually switching between tools.

When NOT to choose this

If you're working offline or in environments with limited internet connectivity, as Roundtable requires access to various AI model APIs.

Tools this server exposes

1 tool extracted from the README
  • roundtable-ai

    MCP server that delegates tasks to specialized AI sub-agents (Gemini, Claude, Codex, Cursor)

Note: The README describes a single MCP server (roundtable-ai) that coordinates sub-agents, but doesn't explicitly list individual tools exposed through MCP. The tool name was extracted from the CLI examples and configuration sections.

Comparable tools

mcp-x-codermulti-agent-codebaseai-workflow-orchestrator

Installation

Installation

Using pip (Standard)

pip install roundtable-ai

Using UV/UVX (Recommended for faster installs)

uvx roundtable-ai@latest

Claude Desktop Integration

Add to ~/.config/claude_desktop_config.json:

{
  "mcpServers": {
    "roundtable-ai": {
      "command": "roundtable-ai",
      "env": {
        "CLI_MCP_SUBAGENTS": "codex,claude,cursor,gemini"
      }
    }
  }
}

FAQ

Does Roundtable AI require separate API keys for each AI model?
No, Roundtable uses your existing CLI tools and API subscriptions with no additional markup.
How does Roundtable handle context continuity between different AI models?
The system packages the initial prompt and relevant context into a 'context bundle' that is passed to each sub-agent, ensuring all participants have the same ground truth without manual copy-pasting.

On Hacker News

Recent discussion from the developer community.

Compare roundtable with

roundtable vs ultimate_mcp_serverroundtable vs mcp-server-chartroundtable vs everythingroundtable vs filesystemroundtable vs time
GitHub →

Last updated · Auto-generated from public README + GitHub signals.