cursor/community-plugins

Plugins from the Cursor community

What it solves

Cursor Directory is a community-driven directory of plugins for the Cursor AI code editor. It provides a centralized place for users to find and submit plugins that enhance the editor's capabilities through rules, MCP servers, skills, and agents.

How it works

The project is a Next.js application that uses Supabase for its database and job queue. When a user submits a GitHub repository URL, the system auto-detects components based on the Open Plugins standard (such as .mcp.json or rules/*.mdc). It then uses a Cursor SDK agent (composer-2) to asynchronously scan the plugin's repository for security verdicts (safe, suspicious, or malicious) before it is listed.

Who it’s for

Developers using the Cursor AI editor who want to discover new community-created plugins or creators who want to share their plugins following the Open Plugins specification.

Highlights

  • Automated Plugin Detection: Auto-detects rules, MCP servers, skills, agents, and LSP servers from GitHub URLs.
  • Security Scanning: Uses an AI agent to automatically review submitted plugins for security risks.
  • ** uma own Job Queue**: Implements an asynchronous scanning process using Supabase Queues (pgmq) and Vercel cron jobs for reliability.
  • Open Plugins Standard: Built to support the Open Plugins specification for interoperability.

Related

  • Project
    AnswerDotAI/gpu.cpp

    A lightweight C++20 interface for GPU compute via WebGPU that simplifies shader dispatch and data movement across native and browser environments.

  • Project
    steipete/oracle

    A CLI and MCP server that bundles project files with prompts to provide grounded AI reviews and audits via APIs or browser automation.

  • Project
    bostrot/wsl2-distro-manager

    A GUI for managing WSL on Windows and native VMs on macOS, featuring an MCP server and AI assistant that can actively operate and configure your Linux environments.

  • Project
    lucidrains/vector-quantize-pytorch

    A PyTorch library implementing various vector quantization techniques, including VQ, Residual VQ, and FSQ, to map continuous data into discrete representations for generative modeling.