agentleFS
Hacck3yHacck3y/ptk/.claude/skills/repo-recap/SKILL.md

repo-recap

Skill2 starsChanged 3 months agoApache-2.0View on GitHub

Generate a comprehensive repo recap (PRs, issues, releases) for sharing with team. Pass "en" or "fr" as argument for language (default fr).

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 1

    1. rtk-ai/rtk★ 87k100% same

    Edited · 0

    None found.

    Wrote their own version · 0

    None found.

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository13 other files it gives its agents.

The file

Show its rules and text

We split this skill into the rules it gives an agent

14 rules. About 0 are common, 8 are distinctive, and 6 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 6

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Check the argument passed to this skill

    line 12
  2. Very specific:

    If en or english → produce the recap in English

    line 13
  3. Very specific:

    If fr, french, or no argument → produce the recap in French (default)

    line 14
  4. Distinctive:

    Before gathering data, verify:

    line 18
  5. Distinctive:

    If either fails, stop and tell the user what's missing.

    line 28
  6. Very specific:

    Run these commands in parallel via gh CLI:

    line 34
  7. Distinctive:

    If this fails (permissions), fallback: authors with write/admin access are those who merged PRs recently.

    line 63
  8. Distinctive:

    No merge conflicts, CI not failing

    line 74
  9. Distinctive:

    If one author has 3+ open PRs, note it as a "cluster" with suggested review order (smallest first, or by dependency chain).

    line 99
  10. Very specific:

    If no releases found, check merged PRs for release-please pattern (title matching chore(*): release *) as fallback.

    line 111
  11. Very specific:

    All PR/issue numbers as clickable links: #123 for PRs, .../issues/123 for issues

    line 128
  12. Distinctive:

    After displaying the recap, automatically copy it to clipboard:

    line 140
  13. Very specific:

    Always use gh CLI (not GitHub API directly, except for collaborators list)

    line 212
  14. Distinctive:

    Keep tables compact — truncate long titles if needed (max ~60 chars)

    line 214

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Repo Recap

Generate a structured recap of the repository state: open PRs, open issues, recent releases, and executive summary. Output is formatted as Markdown with clickable GitHub links, ready to share.

Language

  • Check the argument passed to this skill
  • If en or english → produce the recap in English
  • If fr, french, or no argument → produce the recap in French (default)

Preconditions

Before gathering data, verify:

# Must be inside a git repo
git rev-parse --is-inside-work-tree

# Must have gh CLI authenticated
gh auth status

If either fails, stop and tell the user what's missing.

Steps

1. Gather Data

Run these commands in parallel via gh CLI:

# Repo identity (for links)
gh repo view --json nameWithOwner -q .nameWithOwner

# Open PRs with metadata
gh pr list --state open --limit 50 --json number,title,author,createdAt,changedFiles,additions,deletions,reviewDecision,isDraft

# Open issues with metadata
gh issue list --state open --limit 50 --json number,title,author,createdAt,labels,assignees

# Recent releases (for version history)
gh release list --limit 5

# Recently merged PRs (for contributor activity)
gh pr list --state merged --limit 10 --json number,title,author,mergedAt

Note: author in JSON results is an object {login: "..."} — always extract .author.login when processing.

2. Determine Maintainers

To distinguish "our PRs" from external contributions:

gh api repos/{owner}/{repo}/collaborators --jq '.[].login'

If this fails (permissions), fallback: authors with write/admin access are those who merged PRs recently. When in doubt, ask the user.

3. Analyze and Categorize

PRs — Categorize into 3 groups:

Our PRs (author is a repo collaborator):

  • List with PR number (linked), title, size (+additions, files count), status

External — Reviewable (manageable size, no major blockers):

  • Additions ≤ 1000 AND files ≤ 10
  • No merge conflicts, CI not failing
  • Include: PR link, author, title, size, review status, recommended action

External — Problematic (any of: too large, CI failing, overlapping, merge conflict):

  • Additions > 1000 OR files > 10
  • OR CI failing (reviewDecision = "CHANGES_REQUESTED" or checks failing)
  • OR touches same files as another open PR (= overlap)
  • Include: PR link, author, title, size, specific problem, action taken/needed

Size labels (use in "Taille" column for quick visual triage):

LabelAdditions
XS< 50
S50-200
M200-500
L500-1000
XL> 1000

Format: +{additions}, {files} files ({label}) — e.g., +245, 2 files (S)

Detect overlaps:

Two PRs overlap if they modify the same files. Use changedFiles from the JSON data. If >50% file overlap between 2 PRs, flag both as overlapping and cross-reference them.

Flag clusters:

If one author has 3+ open PRs, note it as a "cluster" with suggested review order (smallest first, or by dependency chain).

Issues — Categorize by status:

  • In progress: has an associated open PR (match by PR body containing fixes #N, closes #N, or same topic)
  • Quick fix: small scope, actionable (bug reports, small enhancements)
  • Feature request: larger scope, needs design discussion
  • Covered by PR: an existing PR addresses this issue (link it)

4. Derive Recent Releases

From gh release list output, extract version, date, and name. List the 5 most recent.

If no releases found, check merged PRs for release-please pattern (title matching chore(*): release *) as fallback.

5. Executive Summary

Produce 5-6 bullet points:

  • Total open PRs and issues count
  • Active contributors (who has the most PRs/issues)
  • Main risks (oversized PRs, CI failures, merge conflicts)
  • Quick wins (small PRs ready to merge — XS/S size, no blockers)
  • Bug fixes needed (hook bugs, regressions)
  • Our own PRs status

6. Format Output

Structure the full recap as Markdown with:

  • # {Repo Name} — Récap au {date} as title (FR) or # {Repo Name} — Recap {date} (EN)
  • Sections separated by ---
  • All PR/issue numbers as clickable links: [#123](https://github.com/{owner}/{repo}/pull/123) for PRs, .../issues/123 for issues
  • Tables with Markdown pipe syntax for all listings
  • Bold for emphasis on actions and risks
  • Cross-references between related PRs and issues (e.g., "Covered by #131")

Empty data handling:

  • 0 open PRs → display "Aucune PR ouverte." (FR) or "No open PRs." (EN) instead of empty table
  • 0 open issues → display "Aucune issue ouverte." (FR) or "No open issues." (EN)
  • 0 releases → display "Aucune release récente." (FR) or "No recent releases." (EN)

7. Copy to Clipboard

After displaying the recap, automatically copy it to clipboard:

# Cross-platform clipboard
clip() {
  if command -v pbcopy &>/dev/null; then pbcopy
  elif command -v xclip &>/dev/null; then xclip -selection clipboard
  elif command -v wl-copy &>/dev/null; then wl-copy
  else cat
  fi
}

cat << 'EOF' | clip
{formatted recap content}
EOF

Confirm with: "Copié dans le presse-papier." (FR) or "Copied to clipboard." (EN)

Output Template (FR)

# {Repo Name} — Récap au {date}

## Releases récentes

| Version | Date | Highlights |
| ------- | ---- | ---------- |
| ...     | ...  | ...        |

---

## PRs ouvertes ({count} total)

### Nos PRs

| PR | Titre | Taille | Status |
| -- | ----- | ------ | ------ |

### Contributeurs externes — Reviewables

| PR | Auteur | Titre | Taille | Status | Action |
| -- | ------ | ----- | ------ | ------ | ------ |

### Contributeurs externes — Problématiques

| PR | Auteur | Titre | Taille | Problème | Action |
| -- | ------ | ----- | ------ | -------- | ------ |

---

## Issues ouvertes ({count} total)

| # | Auteur | Sujet | Priorité |
| - | ------ | ----- | -------- |

---

## Résumé exécutif

- **Point 1**: ...
- **Point 2**: ...

Output Template (EN)

Same structure but with English headers:

  • "Recent Releases", "Open PRs", "Our PRs", "External — Reviewable", "External — Problematic", "Open Issues", "Executive Summary"
  • Action labels: "To review", "Rebase requested", "Split requested", "Trim requested", "CI broken", "Waiting on author", "Feature request", "Quick fix", "Covered by PR"

Notes

  • Always use gh CLI (not GitHub API directly, except for collaborators list)
  • Derive repo owner/name from gh repo view, don't hardcode
  • Keep tables compact — truncate long titles if needed (max ~60 chars)
  • Cross-reference overlapping PRs/issues whenever possible
  • author in gh JSON is an object — always use .author.login

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.