MCP servers that give coding agents real developer tooling: linters, package registries, test runners, language docs and local build commands. 13,951 servers in this category.
Lets an AI client consult a user's digital consciousness for direction and review agent work against stored memories and rules.
$claude mcp add mindupload-sdk-python -- npx -y mindupload
Detects prompt injection, jailbreaks, and data exfiltration in untrusted text before it reaches an LLM.
$claude mcp add safeprompt -- npx -y @safeprompt.dev/mcp
Compiles coding agent sessions and pull request threads into a cited, plain-text knowledge book for your repository.
$claude mcp add crewlore -- uvx crewlore
Predicts SQL query costs, explains execution plans, and suggests optimized rewrites for 12 different database engines.
$claude mcp add anumana -- uvx anumana-mcp
Enables AI agents to perform deep web research using SearXNG for searching, Markdown-based page fetching, and multi-source information gathering.
$claude mcp add mcp-server -- uvx cluefinch
Enables AI agents to list, inspect, and replay recorded LLM calls offline for deterministic debugging without additional API costs.
$claude mcp add agent-observability -- uvx agent-observability-trace-cli
Verifies AI-generated work by checking completion contracts, requirement states, and building release receipts locally.
$claude mcp add release-gate -- uvx dailyaiagents-release-gate
Lets an AI run tests and controlled staging operations on a server using bounded capabilities instead of a general-purpose shell.
$claude mcp add runner-mcp -- uvx aifordable-runner-mcp
Analyze and improve how AI assistants like ChatGPT and Claude cite your business through GEO audits and automated fixes.
$claude mcp add cited -- uvx cited-mcp
Lets an AI client create, debug and deploy Home Assistant automations, dashboards, themes and configs through a safe native API.
Lets an AI client run local chat, RAG over your files, desktop and browser agents, coding swarms, LoRA training, and media generation on your own GPU.
$claude mcp add guaardvark -- uvx guaardvark
Allows an AI client to explore GitHub repository structures and read the contents of specific files within a repository.
Rebuild software from behavior through black-box observation, clean-room reimplementation, and behavioral verification.
A local-first, governed AI agent runtime for Python that provides permissions, memory, and MCP connectivity.
Runs each reasoning step through a team of specialist AI agents to produce multi-perspective analysis and a synthesized answer.
Allows an AI client to monitor long-running coding agent workflows, queue states, proofs, and incident summaries.
$claude mcp add cortexpilot-public -- uvx cortexpilot-orchestrator
Lets an AI client search, create, update, tag, and organize notes and notebooks in a Joplin note-taking database.
$claude mcp add joplin-mcp -- uvx joplin-mcp
Lets an AI client discover, add, enable, and proxy other MCP servers, aggregating their tools under unified prefixes at runtime.
$claude mcp add magg -- uvx magg
Lets an AI client manage Yandex Tracker issues, queues, comments, worklogs, and run advanced query searches.
Lets AI agents manage RTC wallets, send signed transfers, find bounties, publish BoTTube videos, and message other agents via Beacon.
$https://github.com/Ljunqiang/rustchain-mcp/releases/download/v0.4.0-mcpb/rustchain-mcp.mcpb