Glama
MCP server directory with agent-oriented quality scoring.
About
Glama is an MCP server directory that stands out by scoring how well each server's tools describe themselves to AI agents. Its quality score combines tool-definition quality (purpose clarity, usage guidelines, parameter semantics) with server coherence (disambiguation, naming consistency, tool-count appropriateness), graded into tiers from A to F.
The service crawls public MCP repositories automatically, so many servers appear without any action from their authors. Maintainers can claim their listings, add a glama.json manifest declaring maintainers, and trigger manual syncs. Server authors treat the score as a concrete checklist for making their tools more usable by agents.
What it can do
- Automated crawling of public MCP servers
- Agent-oriented tool quality scoring with A-F tiers
- Maintainer claims and manual sync
- Embeddable quality badges for READMEs
Setup
Browse and claim servers at glama.ai; add glama.json to a repo root to declare maintainership.
Prerequisites
A public MCP server repository for claiming a listing.
Use cases
Works with
Trust score — 7/10
Scored 7/10 (metrics tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 92; Trending score 72/100 from researcher assessment.
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.