MCP servers for repositories, version control, CI, error tracking, sandboxes and code execution. Each entry is scanned with the deterministic Capability-Flow Trust Model — grades are computed, never self-reported.
Undo anything your AI agent does. A universal time machine that records every side effect an autonomous agent produces — file writes, shell commands, git state, HTTP mutations — and reverses them with one command.
MCP server for AI Agent Template - Provides structured patterns and context for AI-assisted development
CLI no oficial para UPC Aula Virtual (Blackboard Learn) — acceso desde la terminal y MCP para Claude
MCP server for Build in Public - automatically share your dev progress on Twitter with AI-powered suggestions
MCP server for CheckPJ - Brazilian company data API
[](https://www.npmjs.com/package/context-bonsai-mcp)
Context Engine MCP bridge (http/stdio proxy combining indexer + memory servers)
MCP for debugging with LLDB (C, C++, Rust, Swift) and Python
Enable text-only LLMs to parse and understand frontend animations — MCP server for dynamic UI debugging without vision input
MCP server for generating and parsing Word/PDF/Excel/PowerPoint documents using docxtpl
A code analysis and retrieval system that combines graph databases, vector search, and LLMs to understand and query codebases through natural language.
MCP server for greed-compute — gives LLMs stateful Python execution
MCP server that converts YAML to validated JSON — a safer emission surface for LLMs generating structured JSON payloads.
MCP server integrating a 7-stage agentic frontend workflow: design audit → component gen → E2E test → self-healing CI
MCP server to expose localhost ports via ngsrv tunnels (Cursor, Claude).
Model Context Protocol server exposing queryable Nushell documentation and a Nushell execution environment.
Codex plugin for running isolated Pi coding-agent subprocesses through MCP.
Progressive Insurance MCP connector for personal AI agents
MCP server for interacting with Reddit - browse subreddits, read and create posts, comments, vote, search, and manage messages
ScreenBook — a design workspace for AI agents to present to humans. HTML screens + JSON data in real device frames, with review mode, flows, themes, self-tests, a CLI and an MCP server. Fully offline, no build step, and any MCP agent can drive it.
AI-powered code analysis server for Claude Desktop and MCP clients - risk detection, snapshots, and intelligent code safety
Intranet Steel Browser tester — CLI + MCP (open session & URL)
MCP server for ThinkingEngine (TE) analytics platform — event analysis, SQL queries, dashboards, user segments, and operations
Enterprise Agent Supervisor - Rules Engine MCP Server for AI Agent Governance