Development MCP servers

Read, write, and search local files with sandboxed directory access.

Fetch a URL and return clean Markdown — the canonical web-reading tool.

Issues, PRs, code search, and repository operations across GitHub.

Cross-browser automation — the same Playwright you write tests with.

Read, search, and manipulate local Git repositories.

Issues, merge requests, and project management for GitLab.com and self-hosted.

Drive a real Chromium browser — navigate, click, fill forms, screenshot.

Internal MCP servers from the Cursor IDE team.

Run MCP servers inside Visual Studio Code.

Connect Claude to your JetBrains IDEs (IntelliJ, PyCharm, WebStorm).

Run untrusted code in a secure cloud sandbox, with filesystem and network.

MCP servers for the Windsurf AI IDE.

Search and manage npm packages through MCP.

Headless spreadsheet engine + MCP server for Node.js and AI agents.

Open-source local execution runtime for MCP agents: policy-gated tool calls, scoped approvals and durable receipts.

Read-only validation of rights-aware production metadata manifests for AI video workflows.

Agent-asset exchange: discover, run and trade TAIP/1 AI packs through one MCP.

Universal MCP server for discovering, chatting with and registering agents via the HOL Registry Broker.

Open-source Model Context Protocol server for XGR.Network and XDaLa workflow and blockchain infrastructure.

Persistent memory, semantic search, and shared project context for AI agents with traceable sources and scoped access.

Delegate exact AI tasks to coworkers through Wagglet when you hit your subscription limit.

Give your coding agent a live URL: publish sites over HTTPS and deploy always-on agents.

Local-first memory MCP for coding agents: SQLite + FTS5 + vectors, no API key, no cloud.

MCP server for shared AI-agent troubleshooting and communication

Code security scanner MCP server - scans secrets, dependencies, and insecure code patterns