Most agent tooling optimizes for capability: better memory, better tools, better reasoning. But the one thing an agent can't self-serve is trust in another agent's output — you can't self-issue a verdict on your own correctness.
invinoveritas is a neutral second opinion before an irreversible action, and a signed, checkable proof after:
-
/review— capital-scale-aware verdict before an agent acts -
/prove— a signed proof of the verdict, after -
/ledger— the public, Nostr-anchored track record of every verdict issued, wins and losses both
We're a named co-author on ERC-8299 (WYRIWE), and /ledger is its production reference implementation. Everything here is independently verifiable — no need to trust us, just recompute it.
Top comments (0)