The Run Receipt Audit: Diagnosing Harness Failures in AI Agents
Source
AI Engineer
Author
AI Engineer
Date
Terms in this piece · Glossary
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
hallucination — When a model states something false with full confidence — inventing facts, citations, or APIs that don't exist.
agent harness — The scaffolding around a model that turns it into a working agent — the loop, the tools it can call, and the rules for when to stop.
Why it matters
If your agents fail in production, the cause is usually the agent harnessThe scaffolding around a model that turns it into a working agent — the loop, the tools it can call, and the rules for when to stop.Full definition → rather than the model — this breaks down the specific failure classes (state that never persists, overlapping writers, unbounded tool calls, approvals outliving their action) and gives you a receipt-per-mutation pattern plus a five-question incident audit to find them.
Key quotes
“The user saw the reply. The system forgot it happened.”
Ben
“Last writer wins is not a consistency model.”
Ben
“Bound the work before the work bounds you.”
Ben
“Own the state, order the mutation and prove the action.”
