agentic-trust-layer.md

Trust Layer for Agentic AI

A governance plane between AI agents and a personal knowledge base — append-only ledger, human-ruled writes, crypto-shred redaction. Deliberately underdescribed while the feasibility probe runs.

PythonSQLiteEvent sourcingHuman-in-the-loop review

What it is

AI agents accumulate knowledge about the person who runs them — and nearly every memory system ships with the agent deciding what gets remembered. This is the inverse: a governance plane where every write is a proposal, every acceptance is a human ruling, and the ledger records both sides of that exchange. The machine proposes; the owner disposes; the record survives either way.

The invariants I'll name

  • Append-only event ledger. State is a fold over events, never a stored column. Same spine as everything else I build.
  • Human-ruled writes. Nothing enters the knowledge base on a model's word — the same rule my claim ledger applies to politicians, applied to my own agents.
  • Crypto-shred redaction. Deleting something means destroying its key. The hash chain stays intact; the payload is gone everywhere at once, backups included.

There are more invariants than this. They're the interesting part, which is why they're not on this page yet.

Status

Run to production discipline from day zero: the design survived an adversarial review — reviewer, devil's advocate, security, in separate contexts — and implementation is gated behind a time-boxed feasibility probe that has to earn the build plan. If the probe fails, this page becomes a post-mortem. Either way it stays honest.