Failure traces to regression tests

Make every agent failure leave behind a test.

Inspect a failed support-agent trace, isolate deterministic failure modes, propose bounded patches, and turn the incident into permanent regression checks.

Try the fixture No sign-in · precomputed from the tested fixture

One product.

One decision.

Visible evidence.

Interactive proof

Reveal the verified fixture.

This exact five-event JSON trace is evaluated by product.py.

Failed agent trace

[
  {"seq":1,"type":"context","customer_id":"CUS-1042","verified":true},
  {"seq":2,"type":"tool","name":"lookup_customer","args":{"customer_id":"CUS-1042"}},
  {"seq":3,"type":"context","customer_id":"CUS-1402"},
  {"seq":4,"type":"tool","name":"issue_refund","args":{"customer_id":"CUS-1402","amount":750}},
  {"seq":5,"type":"tool","name":"issue_refund","args":{"customer_id":"CUS-1402","amount":750}}
]

The tested result is ready to inspect.

Reveal the expected decision generated by the product's deterministic acceptance fixture.

How it works

Small surface. Clear chain of proof.

Reconstruct

Order observations, decisions, tool calls, and side effects into one timeline.

Localize

Apply explicit detectors to each supported failure mode.

Prevent

Propose a scoped patch suggestion and an assertion that fails on the original trace; rerun remains NOT_RUN.

✓ Validated sequence✓ One test per failure✓ Proposed patch scope