54% of MCP tools have solid code but zero users. They're invisible to AI agents. Is yours one of them? Paste your GitHub URL → see your grade in 5 seconds. Free. No signup. We scored 39,762 MCP servers. You get a clear grade (F to B+), a badge for your README, monthly ecosystem reports, and transparent methodology. Open source (MIT). Built for builders who want to know if their tool is any good.
No reviews yetBe the first to leave a review for Agent Tool Intelligence
Maker
📌
When we first scored 39K MCP servers, 85.7% got Grade B — useless for differentiation. We rebuilt from scratch with an additive
model combining code quality + community signals + real usage data. The result: 6 grades spread naturally from F to B+.
A/A+ are reserved for tools with both quality AND community adoption — none qualify yet. That's honest.
Happy to answer questions about the scoring methodology or what we learned from 39K tools.