MCP server directory

WhatsApp Business API template management via Meta Cloud API. Create, validate, and send all 12 template types (text, image, video, document, location, authentication, carousel, coupon, catalog, MPM, limited-time offer, and flows) from any MCP client.

An MCP server (and command-line tool) to provide a dynamic map of chat-related files from the repository with their function prototypes and related files in order of relevance. Based on the "Repo Map" functionality in Aider.chat

Structured planning, parallel execution in git worktrees, and deep validation for Claude Code. Turns a one-line objective into a validated DAG of modules executed by worker agents, each self-checked and cross-module-reviewed before merge-back. 7 MCP tools: `validate`, `validate_plan`, `memory_recall`, `memory_save`, `iteration_state` (per-run scoped, with stagnation/velocity/oscillation detection), `forge_logs`, `session_state`. Stdio-only. Zero telemetry.

MCP server for [OpenClaw](https://github.com/openclaw/openclaw) AI assistant integration. Enables Claude to delegate tasks to OpenClaw agents with sync/async tools, OAuth 2.1 auth, and SSE transport for Claude.ai.

Multi-LLM deliberation with anonymized peer review. Runs a 3-stage council: parallel responses → anonymous ranking → synthesis. Based on Andrej Karpathy's LLM Council concept.

Deterministic multi-agent orchestrator for 37 CLI coding agents (Claude Code, Codex, Cursor, Aider, Gemini CLI, GitHub Copilot CLI, OpenHands, OpenCode, Goose, Qwen, Ollama, and more). MCP server mode (stdio + HTTP/SSE) exposes the orchestrator to any MCP client. Git worktree isolation per agent, HMAC-chained audit trail, cost-aware model routing via contextual bandit. Apache 2.0.

Adversarial test generation for AI coding sessions. Detects language and framework; writes tests; runs them; classifies failures via R12 (real_bug / environment / test_bug). Ships 8 adversarial scenario categories. Works in Cline directly and in 8+ editors via Cline host coverage. 162 plugin tests. MIT.

🐍 🏠 ☁️ — Service-boundary-aware document exchange center for coordinating heterogeneous LLM code agents. Versioned Markdown store, pub-sub notifications, diff-aware updates, and lifecycle stage tracking. Each service registers as a sub-project; agents subscribe to cross-service document changes and receive structured diffs via MCP.

Manage a 5dive agent fleet from any MCP client: file tasks, message and inspect agents, and read the fleet digest. Thin adapter over the 5dive CLI.

Execute any LLM-generated code in a secure and scalable sandbox environment and create your own MCP tools using JavaScript or Python, with full support for NPM and PyPI packages

📇 ☁️ 🏠 🐧 🍎 🪟 – Code-mode MCP server: collapses N user-defined tools into a two-tool surface (`execute_code` + `docs_search`). Backed by a unified capability manifest (`allowedHosts` / `cpuMs` / `memoryLimitBytes`) honoured across three sandbox kernels — in-process node:vm, WASM (QuickJS / Pyodide / Wasmtime), and remote microVM (E2B / Cloudflare Sandbox). Token-saving benchmark in repo CI.

Connect AI agents to MATLAB — execute code, run async jobs with progress reporting, get interactive Plotly plots, expose custom .m functions as tools, and monitor via live dashboard.