Hlido MCP server

What it does

Hlido exposes an independent AI-agent review registry over MCP, providing trust scores, evidence-backed scorecards, and claim audits for AI agents. It is for developers and AI systems that need to verify the trustworthiness and capabilities of other AI agents before use or integration.

• Query trust scores and tier verdicts for a specific AI agent to decide whether to use it.
• Search the registry to find trusted agents matching a particular need or capability.
• Verify specific claims about what an agent can do by retrieving per-claim PASS/FAIL audit results.

Tools this server exposes

Tool	Does
trust_check	Retrieve the trust score, tier verdict, and evidence for a given agent.
find_trusted	Search the registry for trusted agents matching a specified need or capability.
verify_claim	Check whether a specific claim about an agent's functionality is audited as PASS or FAIL with evidence.
compare_agents	View side-by-side scorecard comparison of multiple agents.
get_scorecard	Retrieve the full sanitized scorecard JSON for an agent.
find_similar_agents	Find semantically similar agents to a given agent.
submit_agent	Nominate an agent for independent review by Hlido.
report_review_issue	Flag a problem or inaccuracy with a published review.
request_quick_audit	Request a fast re-check of a stale or potentially outdated review.

Before you enable it

An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.

This server has not seen a commit in 2 mo. It may still work, but nobody is currently fixing it — check the alternatives before you build on it.