Operational thinking for
consequential agents.
Substantive guidance for the teams who have to explain—and defend—why an agent received authority.
Release assurance · 2026-09-03 · 7 min read
↗02The release artifact enterprise agents are missing
A model score cannot tell a release authority which exact agent configuration was tested, where it fails, or what authority it may safely receive.
From evaluation output to an enforceable deployment decisionSystem evaluation · 2026-09-02 · 8 min read
↗03Why model evals miss agent authority risk
The most expensive failures occur where a plausible model response becomes an unauthorized tool action, stale decision, or irreversible business event.
Test the harness-visible action, not only the generated textEngineering practice · 2026-09-01 · 6 min read
↗Reproducing agent failures from sealed inputs
A finding becomes actionable when an engineer can replay the same world snapshot, seed, event schedule, and dependency behavior—and verify the same trace root.
Make every failure a regression asset