UNPKG

paqad-ai

Version:

Spec-driven development framework — AI agents that think before they type

20 lines (14 loc) • 804 B
## Fix Proposals ### Failure {{N}}: {{gate}} > {{ac_id}}, {{short test name}} - **AC:** {{ac_id or "untraced"}} - **Failure category:** {{test-failure | test-error | test-timeout | gate-failure}} - **Anchor:** `{{file}}:{{line}}` - **Observed vs expected:** {{short excerpt from failure message}} - **Root cause hypothesis:** `{{file}}:{{line}}`, {{one-sentence cause}} - **Proposed fix:** {{smallest change; never weakens the test}} - **Risk if applied:** {{low | medium | high, single line}} - **Confidence:** {{high | medium | low}} ### Failure {{...next or combined}} Total failures: {{N}} | Combined into {{M}} proposals | High-confidence: {{H}} | Defer to human: {{D}} <!-- If the evidence has no failures, replace this entire body with: --> <!-- Fix Proposals: none, verification passed. -->