Field Notes
The First Real Delegation: An Agent, a Worktree, and a Judge With No Key
A real agent can work in a clean worktree without a signing key. See the delegation boundary that yields a candidate, never a self-issued PASS.
Tag:
Agent Safety
7 articles
Field Notes
I Forked My Agent Harness So It Couldn’t Grade Its Own Homework
If your agent can author its own success, your checks are decoration. Learn the structural split that keeps a harness from grading itself.
Governing Agent Labor
A Report Is Not the Artifact
A session report describes a run. Re-run the command against the artifact on disk before you let that description decide anything.
Field Notes
Six Roads to a False PASS (All One Root Cause)
A valid credential is not proof of what it was about. Learn how six false PASS paths exposed one weak observation boundary.
Field Notes
59 Refusals, Zero Tests: What Mutation Testing Found
Mutation testing exposed 59 refusal paths no test executed. Learn how to measure the safety net instead of trusting a green suite.
Governing Agent Labor
If You Never Fetched It, You Never Researched It
A citation can be fabricated from memory. Fetch and record the source bytes before you let a design claim borrow their authority.
Governing Agent Labor
Why AI-Built Software Needs an Accountability Apparatus
Better models don’t remove the need to bound an agent’s authority: they increase it. The case for treating AI as delegated labour that needs an accountability apparatus outside itself.