mozg

Self-hosted knowledge brains for AI agents, each scored by an exam and honest about its own gaps.

Webfreemium

mozg builds searchable knowledge 'brains' for AI agents from a documentation URL, and measures each brain's coverage with an exam that's re-sat after every update, so a stated accuracy score is a verified fact and gaps are listed publicly rather than hidden. Agents query a brain over MCP with server-side search, and can hand back new findings as owner-approved proposals rather than editing it directly. It is available self-hosted via Docker or as a hosted cloud service, aimed at developers who want an AI agent's knowledge base to be measured and honest about what it doesn't know.

Website

Full match profile

Behind the summary, Matchbox keeps a richer profile of mozg - the signals our matcher actually reads to decide when to surface it. It stays private; claim the listing to see and control it.

Problem & pain-point mapping
Who we surface it to (audience fit)
What it's a strong alternative to
Trust & credibility signals

Claim mozg to unlock →

Try Matchbox with your own problem

Describe what is not working - we’ll show you whether mozg (or something else) actually fits.

Try a problem