agentsclimarketplace

Baton triage

Skill SOSkr/baton/skills/baton-triage

Work-item lifecycle over a board — backend-agnostic CLI + Claude Code skills (GitHub Projects, Plane, ...)

Install
npx -y skills add SOSkr/baton --skill baton-triage

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • 13 days oldThe repository was created 13 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
  • 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

Review/triage a work-item for viability, value, and fit; score it and post the verdict. Use when the user says "triage X", "review idea/item", "evaluate X", "revisar idea". Does not change the stage — records the assessment.

SKILL.md

2.0 KB, 460 tokens by cl100k_base, as published. Nobody here has run it

baton triage

Structured assessment of a work-item. baton reads/writes the board; you provide the judgment. Triage records a verdict; it does not move the stage (approval is baton-approve).

Workflow

  1. Read the item + comments (+ any linked doc + relevant project spec):
    baton show <id>            # title, stage, labels, url
    # full thread: use the backend directly, e.g. gh issue view <id> --comments
    
  2. Score against the criteria below (0-5 each). Adapt example checks to the project's domain.
  3. Post the verdict (score table · strengths · concerns · suggestions · recommendation):
    baton comment <id> --body "$(cat <<'EOF'
    ## Review
    ...
    EOF
    )"
    
    For items with a versioned doc, append a ## Review (YYYY-MM-DD) section there and leave a short verdict comment.
  4. No stage change — the review comment is the record of "reviewed, pending decision". If rejected, use baton-reject.
  5. Present the recommendation (approve / revise / reject). If the user approves on the spot → continue with baton-approve.

Criteria (0-5 each)

  1. Clarity — specific user story, well-defined problem.
  2. Viability — implementable with the current architecture; deps identified; realistic complexity.
  3. Value / Impact — real problem; improves the product's core goals.
  4. Consistency — aligned with existing patterns; doesn't contradict current behavior. Check related items for conflicts/synergies (baton list --label ...).
  5. Completeness — clear acceptance criteria, edge cases, concrete proposal.

Recommendation by score

  • ≥ 15/25 → approve · 10-14 → revise (specific improvements) · < 10 → reject.

Always explain WHY, not just the number.

What ships with it: 1 file

22 B alongside SKILL.md

Keep looking

Skills are one crate of 326,984. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.