Interlock for JavaScript and TypeScript agents: check every tool call before it runs and every result before the model reads it.
WorkerKit CLI: manage your AI workers and browse the kit directory from the terminal
ALLOW_TOOLS/DENY_TOOLS for an MCP server: prefix globs, a curated preset, and error messages that name what exists
VigilSutra AI: open-source, offline-first security and privacy auditor CLI plus AI agent skills. Finds secrets, CI/CD, Docker, Kubernetes, IaC, code and AI-agent risks with redacted, evidence-based SA
Drive the Godot engine from an AI coding agent: build, run, play, and verify your game live.
MCP server that helps AI agents explore OpenAPI specs, search endpoints, and generate TypeScript types.
AI toolkit for hazejs: one API for OpenAI, Anthropic, Gemini, Azure, Bedrock, Cloudflare and OpenAI-compatible providers — text, streaming, tools, structured output, embeddings, agents, images, speech
MCP server exposing Obsidian vaults to remote clients (Claude Code, Claude Desktop) with OAuth discovery and path-level access control
CLI for Release Notes Index — the changelog & release-notes registry for developers and AI agents
MCP server for Sharpe crypto derivatives and market data API
Typed TypeScript client for the 3ngram REST API.
Command line and stdio MCP server for reading, reviewing, and redlining .docx files with folio: tracked-change edits, comments, and comparisons written atomically with version preconditions.
Mojulo — a 3D compiler for coding agents: an MCP server that runs wherever your agent runs, on your machine or in the throwaway Linux box it gives itself (the Claude app or web, ChatGPT work mode with
GitHub for AI agents — repos, files, code search, issues, PRs, commits. Read-only default.
Official MCP App example. Model business scenarios and watch the projections update as you change inputs.
JetBrains IDE plugin giving coding agents programmatic control of the debugger.
Tool definition schemas for the Control Plane: ownership, executors, risk classes, approval modes, idempotency, retry policies, and limits.
WaveSpeed MCP server — run any model on the live WaveSpeed catalog (image, video, audio, 3D) with schema introspection, local-file upload, and price quotes.
MCP server for finding eligible GitHub issues, guiding focused fixes, and preparing pull requests with explicit approval.
MCP server for documentation from local filesystem
Read-only Git inspection for AI agents — log, show, diff, status, blame, grep. No writes.
Model Context Protocol implementation for TypeScript (SEP-834 fork with loosened JSON Schema restrictions)
MCP server for Multiple CLI coding agent integration (Claude + Gemini + Codex)
Post a failing test; a coding agent solves it. Test money is held first, released when you accept.
Local-first safety and verification layer for coding agents: workspace confinement, command allowlists, scope-violation detection and auditable task evidence.
Appo CLI: turn a website into native iOS and Android apps and publish them on the App Store and Google Play, from your terminal or an AI agent.
Self-hosted server that captures design patterns, conventions and decisions from git history and developer feedback and serves them to coding agents.
The framework for production voice agents. Dynamic prompting with full control, in plain code you own. Your coding agent can build, test, and simulate conversations before it ships.
Viator Partner API MCP server for Claude — search tours, activities and experiences, product details, availability schedules, attractions, and destinations. Developed and maintained by AI (Claude Code
Resy MCP server for Claude — developed and maintained by AI (Claude Code)
Hardened filesystem for AI agents — jailed root, read-only default, traversal/symlink-safe.
Official MCP App example. An interactive retention heatmap for cohort analysis.
Easily spin up Model Context Protocol (MCP) servers using Fastify
Render DGMO diagrams to PNG or SVG from the terminal, and wire the DGMO MCP server into AI editors.
Visual feedback for coding agents: stakeholders drop pins on a live app, and the agent reads each pin's selector, screenshot, DOM snippet and thread to fix it.
Model Context Protocol server for vitest-agent. Exposes 30 tools for agent access to test data, TDD lifecycle, and session management.
Annuaire complet
IDE, git, linting et analyse de code. 8,526 serveurs, les plus utilisés en premier. Chacun ouvre la page de l'éditeur.