EvidenceTriage
Curated, sanitized export of 21 agent-skill packages for Claude Code and Codex, gated by an automated publication audit (no secrets, no local paths).
npx -y skills add mj-deving/pai-skills --skill EvidenceTriageAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Turn raw failures into high-signal beads. USE WHEN logs, repro commands, or observed behavior need to become self-contained evidence-first tasks instead of vague TODOs.
SKILL.md
1.1 KB, 216 tokens by cl100k_base, as published. Nobody here has run it
Beads Evidence Triage
Use this skill when the hard part is converting observations into a bead that another agent can execute without archaeology.
Outcome rule:
- facts first
- hypotheses labeled as hypotheses
- acceptance criteria that can disprove the theory
Workflow
- Read references/evidence-bead-template.md.
- Extract:
- exact repro or trigger
- observed evidence
- likely fix surface if justified
- risk or impact
- Create the bead with:
- description for the task and why it matters
--contextfor files, commands, and current state--notesfor provenance usingSOURCES:andkn entry:
- If the work is clearly multi-step, decompose immediately into child beads.
Anti-patterns
Do not create beads that say only:
- investigate this
- fix later
- flaky maybe
If the evidence is weak, say so explicitly instead of pretending certainty.
What ships with it: 2 files
1.4 KB alongside SKILL.md
agents/
- openai.yaml291 B