AI Agent Hub
Back to skills
🤖

PR Blocker Summarizer

AI Agent Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Please follow https://skillhub.cn/install/skillhub.md and install @user_3c6cb52e/pr-blocker-summarizer-sh-nl3poh.

About this skill

Problem

When many pull requests are open, the useful standup signal is narrow: which PRs are stuck on failing checks, changes_requested, or staleness, and which are ready to merge. This skill turns a JSON export of open PRs into a blockers-first digest, rather than listing titles generically.

It expects fields such as title, state, checks, review, and age_days. Missing fields are handled conservatively: a PR is treated as ready only when the evidence is clear, instead of assuming unknown states are unblocked.

How It Works

The skill:

  • Reads a JSON PR list: it does not call the GitHub API and works on exported data.
  • Classifies PRs: flags failing checks, review changes requested, stale or pending review, and merge-ready candidates.
  • Outputs a summary JSON: produces a structured blockers-first digest plus a one-line standup count.

It is a read-only summarizer: it does not merge, comment, or modify PRs. If the workflow needs automatic comments, reminders, or live GitHub data, pair it with other skills.

Use Cases

  • Before standup, convert a JSON export of open PRs into a blockers-first digest for an agent to post.
  • During review backlog triage, identify PRs stuck by failing checks, requested changes, or staleness.
  • Sync owners on which PRs are ready to merge and which are blocked by reviews or checks.
  • Summarize an exported PR list read-only, without calling GitHub APIs or modifying PRs.

Best For

  • Engineers who own standup sync need PR status organized into a blockers-first digest.
  • Engineering leads managing review backlog need to locate failing checks, changes requested, and stale PRs.
  • DevOps engineers using agent daily reports need a one-line summary from PR exports.
  • Authors maintaining example-skill validation flows need a small skill to show pipeline and eval rollout.