Latest MCP Servers

A local‑first, agent‑native control plane that lets large language models drive ComfyUI: generate images, video and audio, author and edit workflows in natural language, manage models and custom nodes, and run the full ComfyUI graph from any LLM provider.

Provides AI agents with a composable toolbox of mathematical operations, delivering inspectable artifacts and independent verification for conjectures, counterexamples, exact computation, and formal proof.

Provides AI‑assisted pull‑request reviews that combine hybrid retrieval‑augmented generation, a structural code graph, and Claude Code to deliver inline, grounded comments with whole‑repository context.

Provides AI‑native PostgreSQL, REST APIs, user authentication, content‑addressed storage, static‑site hosting, and serverless functions, all paid with x402 USDC on Base without user sign‑ups.

Install, test, publish, and manage AI agent skills and MCP servers across dozens of agents via a zero‑dependency command‑line interface.

Provides a structured, version‑controlled vault of markdown documents that AI agents can query through a CLI or an MCP server, delivering cheap, auditable context and governance for solo developers, teams, or enterprises.

Provides an infinite canvas for learning where any selected text can be queried and the answer appears as a fully‑rendered child document, working with Claude Code, Codex, or any MCP‑compatible agent.

Provides a tiled, multi‑host terminal interface that runs entirely in a web browser, streaming real PTYs from any machine over WebSocket and allowing persistent sessions, AI agent interaction, and full‑screen TUI control.

Enables AI assistants to manipulate SideFX Houdini via natural‑language commands, exposing hundreds of tools, resources, and workflow prompts through the Model Context Protocol.

Provides a live sidebar, instant ER diagrams, a powerful CLI, and an MCP server that statically parse Django models to visualize the schema, enforce lint rules, generate factories, and answer AI agents without needing a running Django project or database.

Provides a suite of ArcKit enterprise‑architecture governance commands, agents, and MCP server integrations for the OpenAI Codex CLI, enabling automated architecture workflows, research, compliance, and documentation generation.

A quality‑ranked directory of AI coding skills, plugins, MCP servers and agents for Claude Code, Codex, Gemini CLI and other platforms, enabling rapid discovery and integration.

Provides local, privacy‑preserving indexing and semantic search across code repositories, research papers, and HuggingFace datasets for AI agents.

Provides local‑first access to user‑authorized Google Health API v4 data—including Fitbit and Pixel Watch—through OAuth, exposing the data to AI agents via MCP without transmitting tokens.

Retrieves Bilibili video subtitles, transcripts, metadata, chapters, popular comments and provides search capabilities for AI agents.

Provides a curated collection of production‑ready Claude Code plugins that extend functionality with custom commands, agents, hooks, and MCP servers.

Provides full Simplenote note management via MCP, allowing Claude Desktop and other MCP clients to read, create, update, delete, search, and optionally encrypt notes on the fly.

Enables AI‑driven development of 3D games that run directly in WebGPU‑capable browsers, built on Godot 4.7.1 with a checksum‑locked patch series, MCP server, and asset pipelines.

Provides a bidirectional AI‑native design engineering platform that turns Figma files or live websites into a compact semantic design graph for code generation, design creation, and verification loops, enabling agents to work without REST rate limits or token quotas.

Enables Claude, Cursor, and other MCP‑compatible applications to generate images, videos, audio, text‑to‑speech, and 3‑D models through PiAPI’s extensive media creation models.

Provides a curated list of Model Context Protocol servers together with ready‑to‑use configuration files, enabling AI assistants to interact with development tools, cloud services, databases, and many other resources.

Provides real‑time tennis scores, player data, fixtures, odds and model win‑probability to LLM agents through a Model Context Protocol (MCP) server.

Enables adversary emulation for AI agents, LLM applications, and MCP servers, letting teams test their AI systems against realistic attack scenarios.

Provides a local‑first graph‑backed memory engine for AI agents, persisting decisions, reasons, and contradictions across sessions without relying on external APIs.

Extracts text, subtitles, metadata, videos, covers, images, and public comments from Douyin, Xiaohongshu, Bilibili and YouTube, with optional GetNote and Volcengine cloud ASR for transcription.

Provides high‑performance desktop control for Windows, macOS, and Linux via a Rust NAPI module, exposing screenshot, mouse, keyboard, clipboard, app and window management tools through the Model Context Protocol.

An automated, self‑updating repository that harvests, scores, and stores AI‑related resources, exposing a queryable knowledge graph and an HTTP gateway for agents to inject contextual information without manual curation.

Provides on‑device persistent memory for AI coding agents, enabling retrieval‑augmented generation, context surfacing, and decision tracking without cloud services or API keys.

Provides CLI tools for Jira issue management, worklogs, sprints, boards, and wiki markup via AI agent skills, supporting both Server/DC and Cloud deployments.

Provides comprehensive integration with the Sauce Labs testing platform, enabling AI assistants to manage devices, test jobs, builds, storage, and tunnels through natural‑language conversations.

Provides a markdown‑native, git‑backed knowledge graph that serves as a durable memory and handoff layer for AI agents via a seven‑verb MCP interface.