Code

MCP server for Type Crafter - generate types from YAML specifications

CSS 失效诊断 MCP 服务器：对齐运行时、构建产物与源码，回答样式为什么没生效

MCP server for LLM agents: check an artifact's ontological integrity (the logical consistency of its concepts, authority, and purpose) and derive a domain ontology from real sources, with every structured claim gated by runtime validation.

Crosscode CLI — an MCP server for real-time codebase sync between coding agents

iota embedded policy engine as a pi coding-agent extension: hash-pinned bless, agents.yaml enforcement for read/write/edit/bash, JSONL decision log

YouTube Client — pure Node.js library and MCP server for reliable transcripts (Shorts + long videos), RAG citation packs, video Q&A, playlist/channel batches, chapters, captions, and search. No Go. No yt-dlp. No Python.

Before your AI agent writes code, Sensei tells it what already exists, what to reuse, and what not to touch.

Transparent proxy for MCP servers — records every AI agent tool call and lets you browse, debug, and replay sessions in a live web UI.

微信 ClawBot ↔ pi 桥接服务（iLink 协议直连）| Bridge service connecting the pi coding agent to WeChat ClawBot via the iLink protocol

Cairn — an AI that walks your system and leaves a trail of tests. Autonomous QA agent (UI today; API/unit/docs planned) powered by Claude/OpenRouter, with self-improvement via Langfuse.

AI 代码变更安全网关 — Agent 修改代码必须走 PR 审查，验证通过后方可合并

Local MCP server that gives AI coding agents deep structural knowledge of a JavaScript/TypeScript codebase.

Record any browser page as GIF or video via MCP

Model Context Protocol (MCP) server that enables your MCP client (e.g., Claude Desktop, Cline, Roo Code, Cursor, Windsurf, etc.) to interact with GitHub repositories using your GitHub personal access token.

Ein Stryker Reporter Plugin, das Mutation Testing Ergebnisse nativ über das Model Context Protocol (MCP) bereitstellt.

Agent-first, provider-neutral multimodal OCR CLI for images, PDFs, URLs, JSON schemas, and agentic extraction with Gemini, Kimi, Muse, and OpenRouter.

A Model Context Protocol (MCP) server that provides systematic thinking, mental models, and debugging approaches for enhanced problem-solving capabilities

A lightweight, token-efficient local MCP server that exposes the full Buefy component documentation to any MCP‑compatible client (Cursor, Claude Desktop, Cline, Windsurf, etc.)

Structured CI failure summaries on PRs + compact test/build logs for AI agents. GitHub Action, CLI, MCP. ~80-95% fewer tokens on failures.

MCP server for code reviews - connects LLMs to GitHub and GitLab to analyze Merge/Pull Requests and provide expert feedback