ghostclaw: Skill for Claude Code

.claude/skills/pr-babysitter/SKILL.md

pr-babysitter is a skill for Claude Code from b1rdmania/ghostclaw. It costs 64 tokens per session (1,233 once invoked), scanned A, original, MIT.

A scheduled GitHub pull-request monitor. A pull request is a proposed code change awaiting review; the monitor checks selected repositories for failed automated checks, unanswered review comments, or long inactivity.

In plain words
What is it for?
Use it to watch repositories on a schedule, find pull requests needing attention, and optionally handle simple fixes such as typos, lint errors, or basic test corrections.
Why use it?
It reduces the need to inspect pull requests manually. It can attempt clearly defined small fixes and notify you when human input is needed.

Skill for Claude Code

Written for Claude Code: installed under .claude/. Also seen: names the AskUserQuestion tool.

This is b1rdmania/ghostclaw's own configuration. It tells Claude Code how to work on ghostclaw itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything ghostclaw configures →

Reuse

Borrowing it

Nothing to install: this file belongs to b1rdmania/ghostclaw. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/b1rdmania/ghostclaw/main/.claude/skills/pr-babysitter/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/b1rdmania/ghostclaw

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for pr-babysitter

README.md
[![agentmods](https://agentmods.dev/badge/skills/b1rdmania/ghostclaw/pr-babysitter.svg)](https://agentmods.dev/skills/b1rdmania/ghostclaw/pr-babysitter)
Your own site
<a href="https://agentmods.dev/skills/b1rdmania/ghostclaw/pr-babysitter"><img src="https://agentmods.dev/badge/skills/b1rdmania/ghostclaw/pr-babysitter.svg" alt="Measured on agentmods" height="20"></a>
Per session 64 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,233 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00064 $0.01233
Opus 5 $0.00032 $0.00616
Sonnet 5 $0.00013 $0.00247
Haiku 4.5 $0.00006 $0.00123

Measured 8d ago against content hash 5522ffa06a0c, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

pr-babysitter scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/pr-babysitter/SKILL.md · 92 lines

How it starts

The opening of the file, as written. The whole thing — 92 lines — stays where its author put it; the contents beside it link to each section on GitHub.

PR Babysitter

Automated PR monitoring that only spawns an agent when there's actual work to do. Uses the scheduler's pre_check feature to run a lightweight gh CLI check first — zero API credits burned when all PRs are clean.

Setup

1. Verify GitHub CLI

gh auth status

If not authenticated, run gh auth login.

2. Ask the user

Use AskUserQuestion for each:

  1. Which repos to watch? Accept either:
    • Specific repos: owner/repo1 owner/repo2
    • All repos: gh repo list --json nameWithOwner -q '.[].nameWithOwner'
  2. How often to check? Default: every 30 minutes (*/30 * * * *)
  3. What to watch for? Default: all of the below. Let user pick:
    • Failing CI checks
    • Unanswered review comments (especially from AI reviewers)
    • Stale PRs (no activity for 48h+)
  4. Auto-fix? Should the agent attempt high-confidence fixes (typos, linting, simple test fixes) or always ask first?

3. Create the scheduled task

Build a pre_check script and prompt based on user choices. Example for watching specific repos:

cat > $GHOSTCLAW_IPC_DIR/tasks/pr-babysitter_$(date +%s).json << 'TASKEOF'
{
  "type": "schedule_task",
  "pre_check": "REPOS=\"owner/repo1 owner/repo2\"; for repo in $REPOS; do gh pr list --repo $repo --state open --json number,title,headRefName,statusCheckRollup,reviewDecision,updatedAt --jq '.[] | select(.statusCheckRollup == \"FAILURE\" or .reviewDecision == \"CHANGES_REQUESTED\" or (.reviewDecision == \"\" and (.statusCheckRollup == \"PENDING\" or .statusCheckRollup == \"\")))' 2>/dev/null; done",
  "prompt": "You are a PR babysitter. These PRs need attention:\n\n{{pre_check_output}}\n\nFor each PR:\n1. Check CI status: `gh pr checks <number> --repo <repo>`\n2. Read review comments: `gh api repos/<owner>/<repo>/pulls/<number>/comments`\n3. If CI fails with a clear fix (linting, type error, simple test fix):\n   - Clone the repo, checkout the branch, fix it, push\n   - Comment on the PR: \"Fixed [issue] automatically\"\n4. If review comments are from AI reviewers and the suggestion is high-confidence:\n   - Implement the change, push, reply to the comment\n5. If it needs human judgement:\n   - Message me with: repo, PR number, what's wrong, what you think the fix is\n\nBe conservative. Only auto-fix things you're confident about. When in doubt, ask.",
  "schedule_type": "cron",
  "schedule_value": "*/30 * * * *",
  "context_mode": "isolated",
  "targetJid": "TARGET_JID"
}
TASKEOF

Read the full file on GitHub · 92 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 8d ago First seen · 92 lines · 64 tokens per session scan A 5522ffa06a0c

Subscribe to this mod's changes

pr-babysitter is a skill published in the GitHub repository b1rdmania/ghostclaw (92 stars, last pushed 4mo ago), licensed MIT. It adds 64 tokens to every session and 1,233 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.