Manifesto · July 2026
Permissions Are Not Authority
Access control tells you what an AI agent can do; it says nothing about what it should do. The case for an authority layer that verifies every agent action against the objective a human approved — and produces the evidence to prove it.
Open specification · v0.4 · July 2026
A Taxonomy of Agent Authority Drift
The vocabulary for the ways agents exceed delegated authority: seven drift classes, the authority-chain model behind them, and how this layer composes with identity, authorization, and intent-conformance work. Open for anyone to use and cite.
In progress
More essays on the authority layer are being prepared, including the recommend→execute crossing, why raw LLM judges leak, envelope drift, and what 100 labeled scenarios taught us. The open taxonomy of agent authority drift is the spine that runs through all of them.