Producthermes-provenance-gate
provenance-gate
A Hermes plugin that audits fabricated user decision statements in memory and skill writes, detecting decision content invented by the agent by comparing it against the user's original input.
Timeline (last 90 days)
Oct 22
A developer built three Hermes plugins that audit three failure modes: repetitive loops when AI agents write to their own memory and skill files, fabricated user decisions, and unclear skill provenance.
Unverified50%