babysit-pr

babysit-pr is a command for Claude Code from acolomba/pi-claude-marketplace. It costs 57 tokens per session (1,448 once invoked), scanned A, original, MIT.

A command for an already-open GitHub pull request, a proposed code change waiting for review. It repeatedly reviews and fixes the changes, then checks the SonarQube quality gate, an automated code-quality check.

In plain words
What is it for?
Run it after opening a pull request to resolve review findings and address SonarQube problems. It does not create a pull request and can be run again when the changes are already clean.
Why use it?
It reduces the manual work of finding issues and waiting for fixes between review passes. It aims to leave the pull request in a state that clears the specified automated checks.

Command for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: mentions subagents; names the SlashCommand tool.

Good fit Run it after opening a pull request to resolve review findings and address SonarQube problems. It does not create a pull request and can be run again when the changes are already clean.

Compare 6 commands from other repositories ↓
Install with agentmods
npx agentmods add commands/acolomba/pi-claude-marketplace/babysit-pr
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Clone the repo
git clone --depth 1 https://github.com/acolomba/pi-claude-marketplace

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 babysit-pr

README.md
[![agentmods](https://agentmods.dev/badge/commands/acolomba/pi-claude-marketplace/babysit-pr/github.svg)](https://agentmods.dev/commands/acolomba/pi-claude-marketplace/babysit-pr)
Your own site
<a href="https://agentmods.dev/commands/acolomba/pi-claude-marketplace/babysit-pr"><img src="https://agentmods.dev/badge/commands/acolomba/pi-claude-marketplace/babysit-pr/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for babysit-pr

Your own site · 80×15
<a href="https://agentmods.dev/commands/acolomba/pi-claude-marketplace/babysit-pr"><img src="https://agentmods.dev/badge/commands/acolomba/pi-claude-marketplace/babysit-pr.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 57 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,448 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 $0.00057 $0.01448
Opus 5 $0.00028 $0.00724
Sonnet 5 $0.00011 $0.00290
Haiku 4.5 $0.00006 $0.00145

Measured 11d ago against content hash e1ed43a28e71, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

babysit-pr 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 11d 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/commands/babysit-pr.md · 63 lines

How it starts

The opening of the file, as written. The whole thing — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Babysit PR

Take an already-open pull request and harden it in two phases: a local review-convergence loop, then a SonarQube pass once CI has analyzed the pushed head. The point is to hand a human reviewer a PR that already clears the automated bars, with the expensive review and fixing done in subagents rather than in this conversation's context.

Invoke this yourself after the PR exists ($ARGUMENTS is an optional PR number; default to the current branch's PR). It is safe to re-run — a clean PR converges to a no-op.

Phase 0 — Resolve the PR

Run gh pr view $ARGUMENTS --json number,headRefName,url,state. If there is no open PR for this branch, stop and say so — this command hardens an existing PR, it does not create one. Record the PR number and head branch.

Phase 1 — Review convergence (local, fast)

Converge the PR's diff to review-clean before Sonar even runs. One review pass is never enough — reviewers miss things, fixes introduce new problems, and the only way to know a fix worked is to review again — so this is a bounded loop:

  1. Scope. Review the PR's changes: the branch's commits versus its base (merge-base with main) plus any uncommitted work. If there is nothing to review, skip to Phase 2.
  2. Review. Run /pr-review-toolkit:review-pr over all applicable aspects. Its specialized reviewers run as subagents. It sorts findings into Critical (must fix), Important (should fix), Suggestions (advisory), and Strengths.
  3. Triage. Separate Critical + Important (actionable now) from Suggestions (advisory — these never block finishing).
  4. Fix, one cause at a time. Address each Critical and Important finding at its root. Fix the actual defect; do not silence the reviewer with a blanket lint-disable or by deleting the test that caught it. After each fix, run the project's checks (npm run check and the pre-commit hooks) so a fix cannot quietly break the build, then commit it atomically. Never --no-verify. Delegate independent fixes to subagents.
  5. Re-review. Return to step 2. The re-review is the point: it confirms the fixes landed and catches anything they introduced.
  6. Finish on advisory-only. When a review returns no Critical or Important findings, make one pass over the Suggestions — apply the ones that clearly improve the code, and note in one line why you leave the rest — then stop. Optionally run the simplify aspect as a final polish; the toolkit is built to run that once a change passes review.

Read the full file on GitHub · 63 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. 11d ago First seen · 63 lines · 57 tokens per session scan A e1ed43a28e71

Subscribe to this mod's changes

babysit-pr is a command published in the GitHub repository acolomba/pi-claude-marketplace (23 stars, last pushed 2d ago), licensed MIT. It adds 57 tokens to every session and 1,448 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.