output: Command for Claude Code

.claude/commands/pr-review.md

pr-review is a command for Claude Code from growthxai/output. It costs 7 tokens per session (2,390 once invoked), scanned A, original, Apache-2.0.

An automated review of a pull request, which is a proposed set of code changes for merging into a project. It checks the changes and reports findings without editing the code.

In plain words
What is it for?
Use it to inspect a pull request's code and claims, assess review categories, and add comments pointing to specific lines when needed.
Why use it?
It helps identify problems in the submitted changes while keeping the review focused on what the pull request actually changed.

Command for Claude Code

Written for Claude Code: installed under .claude/. Also seen: mentions CLAUDE.md; mentions subagents.

This is growthxai/output's own configuration. It tells Claude Code how to work on output 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 output configures →

Reuse

Borrowing it

Nothing to install: this file belongs to growthxai/output. 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/growthxai/output/main/.claude/commands/pr-review.md
Clone the repo
git clone --depth 1 https://github.com/growthxai/output

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-review

README.md
[![agentmods](https://agentmods.dev/badge/commands/growthxai/output/pr-review.svg)](https://agentmods.dev/commands/growthxai/output/pr-review)
Your own site
<a href="https://agentmods.dev/commands/growthxai/output/pr-review"><img src="https://agentmods.dev/badge/commands/growthxai/output/pr-review.svg" alt="Measured on agentmods" height="20"></a>
Per session 7 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,390 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 $0.00007 $0.02390
Opus 5 $0.00003 $0.01195
Sonnet 5 $0.00001 $0.00478
Haiku 4.5 $0.00001 $0.00239

Measured 8d ago against content hash 8baefa028ed0, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

pr-review 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/commands/pr-review.md · 159 lines

How it starts

The opening of the file, as written. The whole thing — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Objective

Review this pull request. Do not modify code.

The PR branch is already checked out (shallow clone: fetch-depth: 1 — no full git history/blame). Use gh pr view, gh pr diff, and the local tree.

Tools: mcp__github_inline_comment__create_inline_comment (with confirmed: true) for line pointers only. Do not post a PR summary comment — CI renders it from your structured output.

Scope

Code: Only what this PR adds, changes, or deletes. Neighboring files are fine for conventions; do not file findings against untouched lines unless this PR newly exposes or depends on that problem. No drive-by nits on legacy code.

PR description: Claims to verify (intent, scope, breaks) — not something to grade. Concise is fine. Empty description → at most Nice-to-have; never Must-fix for a weak writeup. If the body claims X and the diff does not support X → a finding under the matching category.

Review process

Work in this order. Do not emit structured output until step 4.

  1. Draft — Walk Review categories once. Provisional verdict, findings, and category scorecard.
  2. Temporal lens (when triggered) — In-process only (no subagent). Focus only on Temporal TypeScript SDK correctness, alignment, and wiring. No separate Temporal category; fold into normal findings.
    • Trigger if the PR touches sdk/core/**, api/**, or @temporalio/* dependency adds/version changes (manifest/lockfile).
    • Review: workflow determinism / sandbox; activity boundaries; retries; heartbeats; cancellation; signals/queries/updates; payloads; replay / patching / versioning; client/worker lifecycle; APIs match the SDK version in use; workers, clients, task queues, interceptors, sinks, schedules coherent with the SDK model.
    • When unsure, check Temporal TS SDK docs or upstream for the version this PR uses — do not guess.
    • Skipping this step when triggered = incomplete review.
  3. Adversarial challenge — Attack the draft before finishing (not a second category walk, not a scorecard row):
    • Claims the diff does not fully support
    • Half-finished wiring (exported but unused, flags with no path, TODOs that leave behavior broken)
    • Silent compatibility / contract changes under-called
    • Failure modes the happy path ignores
    • Generous PASS category rows — upgrade to FAIL and add a Must-fix finding if the challenge sticks
  4. Finalize — Merge surviving Temporal/adversarial points into normal findings (upgrade to Must-fix when needed). Drop weak speculation. Emit structured output and any inline comments.

Read the full file on GitHub · 159 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 · 159 lines · 7 tokens per session scan A 8baefa028ed0

Subscribe to this mod's changes

pr-review is a command published in the GitHub repository growthxai/output (435 stars, last pushed 3d ago), licensed Apache-2.0. It adds 7 tokens to every session and 2,390 once invoked, about $0.0000 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.