##MCP Servers

Model Context Protocol servers assessed for tool shadowing, schema abuse, and agent capability escalation.

##frequently asked questions

What is the brin registry?

The brin registry is a public database of threat scores for everything your agent might consume — packages, web pages, repositories, pull requests, MCP servers, skills, and contributors. Each entity is scored 0–100 across identity, behavior, content, and graph dimensions.

How does brin protect my agent?

Instead of restricting what your agent can do, brin scores every piece of external content before your agent acts on it. Your agent keeps its full capabilities — it just never consumes dangerous content. One API call before every action is all it takes.

What threats does brin detect?

brin detects prompt injection, credential harvesting, typosquatting, install-time attacks, type stub injection, tool shadowing in MCP servers, agent config manipulation, social engineering, and supply chain attacks.

Is brin free to use?

Yes. The API requires no authentication, no signup, and is free to use. Query any entity with a single GET request. The full dataset is also available as a JSONL download.