MCP servers that give coding agents real developer tooling: linters, package registries, test runners, language docs and local build commands. 13,969 servers in this category.
Lets AI agents read, build, debug, and playtest Blueprints and C++ code directly inside a running Unreal Engine editor.
Diagnoses repositories for AI agent context noise, missing CLI commands, build cache efficiency, and flaky tests.
$claude mcp add dev-doctor -- npx -y dev-diagnosis-cli
Acts as a security proxy for MCP servers to redact secrets, prevent prompt injection, and audit tool usage.
Performs deterministic code-structure analysis to generate task pipelines, module graphs, coupling scores, and feature indices without model calls.
Provides a secure toolkit for Claude Code to manage remote server credentials, deploy files via FTP/SSH, and track live sessions through a LAN dashboard.
Lets AI assistants compile, deploy, and call smart contracts on Kaspa-ecosystem networks like Kasplex and Igra.
Converts compact text notation into MusicXML to generate and edit MuseScore sheet music in formats like PDF, MIDI, and MP3 via CLI or MCP.
Enables multi-agent orchestration for Claude Code using a coordinator-worker pattern with automated reporting and long-term memory.
Lints and rewrites MCP tool JSON Schemas to ensure compatibility with Gemini, OpenAI strict mode, and Anthropic.
Validates MCP server configurations across multiple clients to detect broken JSON, path issues, stdout pollution, and leaked secrets.
Provides a centralized registry for discovering and connecting to MCP servers and A2A agents via a single metadata hub.
Deploy an MCP server with a UI view on Cloudflare Workers using Durable Objects for per-session state management.
Provides a local Windows dashboard to inventory, deduplicate, upgrade, and clean up MCP servers and Skills across multiple AI clients.
Runs LLM-generated code in a sandboxed Deno subprocess with restricted permissions to enable Cloudflare Code Mode in Node.js environments.
Diagnoses MCP server failures by probing stdio and HTTP servers to identify handshake, timeout, or authentication issues.
Simulates a voice-first smart display host to test MCP servers with speech, visual screens, and voice-based elicitation.
Measures context token costs for MCP servers and lints tool schemas to prevent context bloat and improve model performance.
Speeds up npm-based MCP server startup times and manages background updates to prevent timeouts in AI clients like Claude Code.
Reduces LLM context usage by allowing models to execute JavaScript programs to interact with multiple MCP servers through a single tool.
Generates consistent ChatGPT and Claude plugin packages from a single specification file for MCP servers.