118,140 servers from the official registry, GitHub and npm. Filter by category, transport and language.
Provides an AI-curated, self-updating directory of Model Context Protocol servers discovered from GitHub and the official registry.
Enables AI agents to build, audit, and manage small-business websites using a spec-first workflow on GitHub and Cloudflare Pages.
Finds and installs GitHub-based skills, plugins, and MCP servers while performing security scans for vulnerabilities and malware.
Manages a fleet of MCP servers by scaffolding new ones, studying GitHub repositories, and checking fleet health against standards.
Provides a self-hosted development environment that allows AI agents to manage work items, access terminals, and interact with repositories via MCP.
Provides a vendor-neutral repository template that gives AI coding agents shared instructions, skills, and MCP guidance to prevent vendor lock-in.
Scans AI-generated GitHub repositories for security vulnerabilities like leaked keys, environment files, and SQL injection.
$claude mcp add --transport http vibescan https://vibescan.bigfloppy.net/mcp
Provides AI editors with code rationale by synthesizing git history, pull requests, and call-graph context.
Allows AI clients to read and update a personal thinking OS stored as JSON in a private GitHub repository via AWS Lambda and Cognito OAuth.
Provides a template for deploying multi-zone MCP groups containing tools, resources, and prompts with integrated guardrails.
Connect AI agents to an AI-native research journal to submit, revise, track, and cite peer-reviewed scientific papers.
An agentic software factory that automates the process from GitHub issue to reviewed draft PR through a multi-agent workflow.
Connects AI editors like Claude Code or Cursor to an autonomous AI pentester to find and prove security flaws in your applications.
Provides a shared working context across multiple AI tools by storing projects, decisions, and handoffs in a private GitHub repository.
Provides guard rails for AI coding agents using written specs, isolated worktrees, and a gated merge queue to ensure code quality.
Connects WhatsApp to GitHub Copilot to read, summarize, and send messages, files, locations, and polls with user confirmation.
Acts as an MCP gateway that allows AI assistants to access GitHub, Linear, Jira, and Cloudflare using individual user OAuth identities without exposing tokens.
Scans MCP servers and tool-using agents for security vulnerabilities, drift detection, and injection paths with SARIF and HTML reporting.
An agentic incident analysis tool that investigates backend service failures by correlating data from monitoring, logs, and repositories.
Validates MCP server manifest files through a GitHub Action to catch common configuration errors before publication.