119,945 servers from the official registry, GitHub and npm. Filter by category, transport and language.
Paste its GitHub repo and we list it after a quick review.
Connects AI assistants directly to production observability data including logs, metrics, traces, exceptions, and alerts.
$claude mcp add last9-mcp-server -- npx -y @last9/mcp-server
Enables AI agents to scrape live websites, bypass anti-bot systems, and extract structured data from any webpage in real-time.
$claude mcp add scrapfly-mcp -- npx -y scrapfly-mcp
Enables AI agents and test suites to perform full-stack introspection and control of Tauri applications via DOM, IPC, Rust state, and SQLite access.
$victauri-cli
Lets an AI agent control macOS by managing windows, applications, and system settings through the Hammerspoon automation engine.
$claude mcp add hammerspoon-mcp -- npx -y @vukvukovich/hammerspoon-mcp
Scores AI-generated UI for genericness using an LLM judge to identify repetitive design patterns and layout novelty.
$claude mcp add slop-eval -- uvx slop-eval-cli
Enables AI agents to interact with the Figma desktop app using the full Plugin API for reading, writing, and executing arbitrary JavaScript.
$claude mcp add localfig -- npx -y localfig
Evaluates agentic models by stress-testing their ability to use MCP tools in real-world environments like GitHub, Notion, and Postgres.
Expose a Kali Linux desktop to AI clients for system monitoring, network reconnaissance, remote control, and Playwright Firefox browser automation.
Enables AI agents to programmatically control macOS via window management, audio, Bluetooth, Spaces, and real-time OS event subscriptions.
$claude mcp add hammerspoon-mcp -- npx -y hammerspoon-mcp
Andrej Karpathy's 2026 AI Systems Engineering Roadmap — 7 production-grade projects covering Context Engineering, Tool Design, Agent Orchestration, Eval Discipline, Reliability Engineering, MCP Servers & Harness Mindset.
An MCP server that offers 15 specialized tools to evaluate and improve the accuracy of RAG based systems.
Agentic reconciliation-break triage system — bounded LangGraph state machine, MCP trust-boundary tool split, human-approval interrupt, and a real eval harness.
MCP-enabled evaluation and claim-grounding toolkit for long-document RAG systems - retrieval evaluation (5 methods), claim verification, a local FastAPI app, a local stdio MCP server for Claude Code/Codex, plus the hoste
Memory and brand-context layer for AI agent systems. Multi-tenant MCP server — 39 tools auto-generated from a live OpenAPI spec. PostgreSQL, per-workspace isolation, scored eval harness.
Governed MCP + Skills agent platform over a legacy engineering-ops system: entitlements, audit, human-in-the-loop writes, eval-gated releases, OpenTelemetry tracing.
Three-agent research system on the OpenAI Agents SDK — Planner/Executor/Reviewer with typed handoffs, a custom MCP server, function tools, input/output guardrails, token-streaming (CLI + SSE), tracing, and an LLM-judge e
Sanitized public extract of a private podcast RAG system in daily use. Hybrid retrieval (BM25 + dense embeddings, RRF fusion) with a 122-query gold-set eval (recall@k, MRR) gated in CI, passages cited to the minute, serv
A support agent for a fictional auto parts store that answers recall, complaint, safety-rating and parts questions from NHTSA's public data with Claude, shows every step on a live system map, and has every answer checked
