Discover MCP Servers

@featured

A sub-harness (both an MCP server and an MCP client). Delegates documentation and source code analysis, as well as interactions with related context-providing MCP servers, to a local or inexpensive OpenAI-compatible LLM. Drastically reduces token usage and context size for coding agents on the top-tier LLM.

@featured

Autonomous native crash debugging for Qwen Code via DAP → MCP. CodeLLDB, crash dumps, runtime evidence and agentic verification.

@featured

Free, 100% local MCP server. Turns your codebase into an AST knowledge graph so AI coding agents (Claude, Cursor, Copilot) get exact caller trees, semantic search, and safe writes — 60% fewer tokens, zero data egress.

@featured

MCP server for VibeKit — build, deploy, and manage hosted apps and chat with their AI agents from Claude Desktop, Cursor, and other MCP clients.

@featured

Self-hosted AI coding agents that live in Slack — run Claude Code or Codex in your own infra, ship PRs, and keep every credential behind a controlled MCP tool layer.