Hi Product Hunt — founder here. Hlido started from a simple frustration: most "AI agent reviews" are paraphrased marketing. So we built a review desk that actually tests agents and publishes the receipts — every vendor claim marked pass/fail/unverified with captured evidence, dated, and re-tested as products change. Scores move in both directions; some of our corrections were downward on agents it would have been more comfortable to leave alone. What you can try today, all free: • Trust checker — type an agent's name, get the verdict: hlido.eu/check • EU AI Act Evidence File — pick the AI tools your company uses, get a printable evidence document (transparency rules start applying Aug 2): hlido.eu/eu-ai-act/evidence-file • MCP server safety register, incidents registry, and everything as JSON/MCP for your own agents to query. Honest limitation: where products sit behind login/payment walls, reviews are public-surface only and labeled with lower confidence — we'd rather tell you the depth honestly than pretend. Nobody pays for a score or placement — that's the entire point. Harsh feedback very welcome.

Hlido

Independent, evidence-backed trust checks for AI agents

Independent, evidence-backed trust checks for AI agents