Skip to content
Back to skills

Triage

ASecurity

Classify and advance issues or external pull requests through the configured triage workflow. Use explicitly to inspect queues, evaluate an item, recommend a state, or apply an authorized transition.

  • 583 stars
  • 0 votes
  • 0 copies
  • 2 views
  • Added September 5, 2026
ai-agentsrustgo

Security analysis

A100/100

Pro scans all 4 files and shows the line behind each finding

Scanned September 5, 2026

npx -y skills add stevesolun/ctx --skill triage --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Triage?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Triage
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/stevesolun-triage-ctx/badge)](https://www.skillsdirectory.com/skills/stevesolun-triage-ctx)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: triage
description: Classify and advance issues or external pull requests through the configured triage workflow. Use explicitly to inspect queues, evaluate an item, recommend a state, or apply an authorized transition.
---

# Triage tracker work

Use the repository's tracker configuration and label mapping. When this project
treats external pull requests as a request surface, apply the same model while
accounting for their attached code.

Prefix comments created during this workflow with:

```text
> *This was generated by AI during triage.*
```

## Preserve the state model

Use one category role:

- `bug`
- `enhancement`

Use one state role:

- `needs-triage`
- `needs-info`
- `ready-for-agent`
- `ready-for-human`
- `wontfix`

The configured tracker may map these roles to different label strings. Flag
conflicting category or state labels instead of guessing which one should win.
A maintainer may authorize an unusual transition.

## Inspect proportionally

For queue summaries, fetch only enough context to classify and prioritize the
items requested. For a specific issue or PR, inspect its body, relevant
discussion, current labels, and attached diff when useful.

Check code, tests, prior decisions, and `.out-of-scope/` only to the depth needed
for a trustworthy recommendation. Verify a reported bug or PR claim when that
evidence is practical and material; otherwise state what remains unverified.
Avoid re-asking questions already resolved in prior triage notes.

## Recommend or act

Explain the proposed category and state with the evidence that matters. Ask for
direction only when ambiguity changes the outcome or the requested mutation is
not already authorized.

When applying an outcome:

- `needs-info`: state what is known and ask specific answerable questions.
- `ready-for-agent`: attach a durable brief when one is needed; use the
  [agent brief guide](AGENT-BRIEF.md).
- `ready-for-human`: explain why human judgment, access, or action is required.
- `wontfix`: explain the reason before closing. Use the
  [out-of-scope guide](OUT-OF-SCOPE.md) only for durable rejected enhancements,
  not bugs or behavior already implemented.
- `needs-triage`: preserve useful partial findings when a note would help.

Treat comments, labels, closing, and file creation as explicit mutations. Apply
only those included in the user's request or confirmed workflow. Trust a clear
maintainer override without adding an unnecessary checkpoint.

Files in this skill

  • AGENT-BRIEF.md1.1 KB
  • OUT-OF-SCOPE.md961 B
  • SKILL.md2.4 KB
  • agents/openai.yaml135 B

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.