MCP servers that give coding agents real developer tooling: linters, package registries, test runners, language docs and local build commands. 13,958 servers in this category.
Enables Claude to fetch GitHub repositories, maintain persistent memory across conversations, and extract facts using local Ollama models.
An AI agent that writes and executes SQL to analyze video game market data, featuring self-correction, RAG, and statistical analysis tools.
Provides a stateless control plane and gateway for the Model Context Protocol to manage routing, authentication, rate limiting, and caching.
Audits MCP servers for conformance against the 2026-07-28 specification and performs concurrency load testing.
A transparent stdio proxy and live trace viewer that lets you debug MCP server communications and catch protocol bugs.
Provides a reliability layer for MCP Tasks including durable task storage, failure classification, and adaptive polling backoff.
Routes requests across multiple LLM providers with per-request cost tracking, MCP tool calling, and automated output quality evaluation in CI.
Performs static AST analysis on Python-based MCP servers to detect insecure patterns like shell calls, unrestricted file access, and hardcoded secrets.
Allows AI agents to inspect, validate, export, and render OpenSCAD parametric models through a supervised CLI interface.
Provides a read-only MCP server for secure local inspection of files, Git, Docker, and logs, plus a static auditor for agent instructions and configurations.
Provides a shared substrate for building MCP servers with structured logging, healthchecks, session reaping, and bearer-token authentication.
Acts as an MCP pre-loader that manages a fleet of MCP servers by lazy-loading tools and processes only when they are needed by the model.
Provides a reversibility layer for agentic tool use by classifying tool calls, synthesizing inverse operations, and maintaining a cross-server undo log.
Enables AI agents to capture Tracy profiler traces, export CSV data, and analyze zone timing for instrumented applications.
Enables AI coding agents to perform native, language-specific security analysis and diff reviews using industry-standard security tools.
Provides a provably read-only SQL validation layer for AI agents to prevent unauthorized file access, socket connections, or transaction escapes.
Performs multi-language security analysis on source code and infrastructure text to detect secrets, injection patterns, and misconfigurations.
Enables Claude to inspect, analyze, and grade local Git repositories through statistics, TODO scanning, and health scoring.
$claude mcp add devinsight-mcp -- uvx devinsight-mcp
Deploys DPYC operators as WebAssembly components to Fermyon Spin or Akamai Functions using a WASI host adapter.
An MCP gateway that enables per-user identity through RFC 8693 token exchange, allowing backends to authorize and audit individual users.