oz-for-oss: Skill for Claude Code

.agents/skills/update-pr-review/SKILL.md

update-pr-review is a skill for Claude Code, Codex from warpdotdev/oz-for-oss. It costs 82 tokens per session (1,334 once invoked), scanned A, original, MIT.

An updater for repository-specific pull-request review guidance. It uses human comments and replies from past reviews to improve the local rules for code and specification reviews.

In plain words
What is it for?
Use it to extract recurring feedback from PR conversations and update local review guidance for code or specification changes.
Why use it?
It helps future reviews reflect repeated reviewer feedback while preserving the shared review contracts.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: installed under .agents/ (shared by several agents).

This is warpdotdev/oz-for-oss's own configuration. It tells Claude Code and Codex how to work on oz-for-oss 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 oz-for-oss configures →

Reuse

Borrowing it

Nothing to install: this file belongs to warpdotdev/oz-for-oss. 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/warpdotdev/oz-for-oss/main/.agents/skills/update-pr-review/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/warpdotdev/oz-for-oss

Made for: Claude Code, Codex.

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 update-pr-review

README.md
[![agentmods](https://agentmods.dev/badge/skills/warpdotdev/oz-for-oss/update-pr-review.svg)](https://agentmods.dev/skills/warpdotdev/oz-for-oss/update-pr-review)
Your own site
<a href="https://agentmods.dev/skills/warpdotdev/oz-for-oss/update-pr-review"><img src="https://agentmods.dev/badge/skills/warpdotdev/oz-for-oss/update-pr-review.svg" alt="Measured on agentmods" height="20"></a>
Per session 82 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,334 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 $0.00082 $0.01334
Opus 5 $0.00041 $0.00667
Sonnet 5 $0.00016 $0.00267
Haiku 4.5 $0.00008 $0.00133

Measured 7d ago against content hash 3a1ba76f8cb1, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

update-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 7d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (scripts/aggregate_review_feedback.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

.agents/skills/update-pr-review/SKILL.md · 100 lines

How it starts

The opening of the file, as written. The whole thing — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Update PR Review

Use this skill to improve the repo-local review companions .agents/skills/review-pr-local/SKILL.md and .agents/skills/review-spec-local/SKILL.md from real reviewer feedback. The shared review-pr skill and the repo-local core .agents/skills/review-spec/SKILL.md are the cross-repo contracts and are read-only from this loop.

The repository uses two separate review skills: review-pr for code pull requests and review-spec for spec-only pull requests (PRs where every changed file lives under specs/). Feedback from each category of PR should be routed to the corresponding repo-local companion.

Write surface

This self-improvement loop may only write to:

  • .agents/skills/review-pr-local/ (and SKILL.md inside it)
  • .agents/skills/review-spec-local/ (and SKILL.md inside it)

It must NOT touch:

  • the shared review-pr skill
  • .agents/skills/review-spec/SKILL.md (the core contract)
  • any file under .github/issue-triage/ (that taxonomy is owned by the update-triage loop)
  • any other core skill

The self-improvement runner enforces this via a git diff check against allowed prefixes before pushing. A violation aborts the run.

Inputs

  • Optional repository override if you are not running from the target checkout.
  • Optional time window override when you need something other than the default seven-day lookback.
  • Optional agent-login override if the review comments were not authored by the default bot identities.

Workflow

  1. Verify GitHub CLI auth:
gh auth status
  1. Aggregate the feedback for pull requests updated over the last week with the bundled script:
python3 .agents/skills/update-pr-review/scripts/aggregate_review_feedback.py

By default this targets the current repo, looks back 7 days, and analyzes review comments authored by the bot identities used by the PR review workflow (warp-dev-github-integration[bot]). It also collects broader human review comments from those PRs so the skill can learn from reviewer norms even when they were not replying directly to the bot. The script writes structured JSON to a temporary file and prints the temp-file path. Treat that file as scratch state for this skill, not as a user-facing deliverable or final output. If you need a repository other than the current checkout, pass --repo owner/name. If you need a different author identity, pass --agent-login <login> one or more times.

Read the full file on GitHub · 100 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 7d ago First seen · 100 lines · 82 tokens per session scan A 3a1ba76f8cb1

Subscribe to this mod's changes

update-pr-review is a skill published in the GitHub repository warpdotdev/oz-for-oss (307 stars, last pushed 5d ago), licensed MIT. It adds 82 tokens to every session and 1,334 once invoked, about $0.0004 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.