All skills
repeat-failure-analysis
FoundationJIT onlyanalysis-agent/task-agent/review-agent: use when repeated failure needs a new hypothesis or proof path; skip an initial failure with verified cause and a different action.
- Group
- Engineering workflow
- Supported roles
- MainAnalysisTaskReview
- Source
- SKILL.md
Use when
Trigger signals that make this skill the right owner
- same path cause patch shape or validator failed twice
- repair repeats an approach contradicted by current task evidence
Do not use when
Anti-triggers — as binding as the triggers above
- first failure has a verified cause and materially different next action
Required inputs
What must be supplied before this skill can decide anything
- current task contract
- selected primary Professional Skill
- task-local trigger evidence
Output contract
What this skill owes the next role
- observed facts, rejected hypotheses, same-pattern impact, next hypothesis, changed execution path, validation method, proof limits, and blocker if any
Escalation signals
When this skill must hand the decision back
- no new hypothesis or safer probe distinguishes another attempt
Loaded by
This skill never owns a task. These professionals may pull it in as Layer 3.
Targeted references
Each reference carries its own load condition. None of them enter context by default.
| Reference | Load when | Do not load when | Required by |
|---|---|---|---|
| repeat-failure-checklist.mddecision-checklist | the same path cause patch or validator has failed twice | this is the first verified failure or the next path is materially different | AnalysisTaskReview |