> **pr-labeling** — skill 1 of 4 in [vercel-labs/skills](https://skillsdocs.com/vercel-labs/skills).
>
> Book (all skills, one file): https://skillsdocs.com/vercel-labs/skills.md
> Machine manifest: https://skillsdocs.com/vercel-labs/skills/.well-known/agent-skills/index.json
> Install the book: `npx skills add vercel-labs/skills`
> Upstream: https://github.com/vercel-labs/skills/blob/main/.automations/pr-labeling/skills/pr-labeling/SKILL.md @ `main`
> Raw bytes, no header: https://raw.githubusercontent.com/vercel-labs/skills/main/.automations/pr-labeling/skills/pr-labeling/SKILL.md
> Base for relative paths: https://raw.githubusercontent.com/vercel-labs/skills/main/.automations/pr-labeling/skills/pr-labeling/
> Licence: MIT — https://spdx.org/licenses/MIT.html
>
> Content © its authors, served unmodified. Takedown: https://github.com/DreambaseAI/skillsdocs/issues/new?labels=takedown&title=Takedown+request

<!-- Verbatim upstream SKILL.md follows, YAML frontmatter included. -->

---
name: pr-labeling
description: Classify a skills CLI pull request from its description and patch, while respecting existing labels.
---

# Label a skills CLI pull request

1. Use `read_pull_request` with the number in the trigger message. Read the title, description,
   current labels, and changed file patches. If the head commit no longer matches the triggered
   commit, stop so a newer run can classify it.
2. If the pull request already has `bug`, `documentation`, or `enhancement`, stop. Keep labels
   chosen by maintainers or earlier runs.
3. Choose one label from the primary change supported by the patch:
   - `documentation`: changes only documentation, examples, or explanatory text, without changing
     CLI behavior.
   - `bug`: corrects existing CLI behavior, including installation, update checks, source parsing,
     agent detection, or security checks.
   - `enhancement`: adds or intentionally expands CLI behavior, commands, agent support, or
     integration support.
4. If the patch is truncated, mixed, or insufficient to classify confidently, stop without adding
   a label. Do not infer a label from the title alone.
5. Use `add_label` once, with the triggering pull request number and the chosen label.