Adversarial verification for AI agents: an independent skeptic attempts to refute a claim before the agent acts on it, returning a structured verdict (refuted / supported / inconclusive) with calibrated confidence, findings, citations, and an Ed25519-signed receipt on paid tiers.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Not distributed through a package registry we index.
Attributed to the source that supplied each field. Treated as claims, not facts.
No long description beyond the summary.
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.
Shared memory for all your AI agents, your whole team and every MCP client — save, search, recall.