hotato failure record · FAIL · refund-claimed-not-issued
refund-issued failed: expected a refund.create tool call; none was found in the trace
| Dimension | Status | Observed |
|---|---|---|
| Outcome | FAIL | expected a refund.create tool call; none was found in the trace |
| Policy | PASS | The policy assertion passed against the supplied evidence. |
| Conversation | PASS | The latency assertion passed against the supplied evidence. |
| Speech | INCONCLUSIVE | no timing context was provided for a latency field |
| Reliability | NOT_RUN | No assertion ran in this dimension. |
Gate authority
Deterministic gate: FAIL · policy all_deterministic_assertions_pass
Model advisory: UNAVAILABLE · gate not enabled
Reliability
pass@1=0.400 pass@k=1.000 pass^k=0.000 (2 of 5 trials passed); 95% Wilson interval [0.117621, 0.769280]
Primary assertion
assert.tool_call · FAIL
Expected: The declared tool_call conditions hold against the supplied evidence.
Observed: expected a refund.create tool call; none was found in the trace
Evidence: assertion-refund-issued-evidence
Evidence references
| Reference | Kind | Locator | Digest |
|---|---|---|---|
assertion-refund-issued-evidence | tool_call | (digest only) | sha256:8adec3cb69c1c... |
assertion-disclosure-present-evidence | policy_match | (digest only) | sha256:4a929d631fa1c... |
assertion-yield-latency-evidence | timing_measurement | (digest only) | sha256:4bfed19f36210... |
assertion-speech-latency-evidence | timing_measurement | (digest only) | sha256:ad92320c255d3... |
Reproduce
hotato record render source-result.json --out record