review-pr

review-pr is a skill for Claude Code, Codex from authgear/authgear-server. It costs 72 tokens per session (4,502 once invoked), scanned A, original, Apache-2.0.

A structured review of changes on the current Git branch that are not yet in the upstream main branch. It includes a second pass to verify each finding.

In plain words
What is it for?
Checking a pull request's purpose, API changes, code quality, bugs, security, and performance.
Why use it?
It reduces the chance of reviewing the wrong changes or reporting an issue without checking it objectively.

Skill for Claude CodeCodex

About the project

Authgear is an open-source identity and authentication service that provides login, account management, multi-factor authentication, passkeys, and single sign-on for applications. SaaS and mobile-app developers can run it themselves or use Authgear Cloud to manage consumer and enterprise user access. The catalogue entries relate to operating and configuring Authgear's authentication workflows.

authgear/authgear-server · 2,024 stars · on GitHub

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.

agentmods
npx agentmods add skills/authgear/authgear-server/review-pr
Any agent
npx skills add authgear/authgear-server --skill review-pr
Clone the repo
git clone --depth 1 https://github.com/authgear/authgear-server

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 review-pr

README.md
[![agentmods](https://agentmods.dev/badge/skills/authgear/authgear-server/review-pr.svg)](https://agentmods.dev/skills/authgear/authgear-server/review-pr)
Your own site
<a href="https://agentmods.dev/skills/authgear/authgear-server/review-pr"><img src="https://agentmods.dev/badge/skills/authgear/authgear-server/review-pr.svg" alt="Measured on agentmods" height="20"></a>
Per session 72 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,502 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00072 $0.04502
Opus 5 $0.00036 $0.02251
Sonnet 5 $0.00014 $0.00900
Haiku 4.5 $0.00007 $0.00450

Measured 3d ago against content hash 46937f004e39, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

review-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 3d 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/review-pr/SKILL.md · 101 lines

How it starts

The opening of the file, as written. The whole thing — 101 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Produce a code review report for the commits on the current branch that are not yet on the upstream main branch. This skill has three phases: scoping, an initial report, and a mandatory verification pass that re-derives every finding with an objective check (not re-reading the same code and nodding along).

Phase 1: Scope the PR correctly

Do NOT assume the local master/main branch is up to date — in this repo it is frequently stale by hundreds of commits. Establish the true upstream base first:

  1. Run git remote -v and identify the canonical upstream remote (the one pointing at authgear/authgear-server, conventionally named authgear; ask the user if ambiguous or absent).
  2. git fetch <upstream-remote> main --quiet
  3. git merge-base HEAD <upstream-remote>/main — this is the true base, regardless of what local master points at.
  4. git log --oneline <base>..HEAD to enumerate exactly the commits in scope. Sanity-check this list against git log --oneline -5 from the repo status — if the top commits don't match what the user expects, stop and re-check the remote/branch choice before proceeding.
  5. git diff --stat <upstream-remote>/main..HEAD — if the stat includes files unrelated to the commit subjects (e.g. translation files, unrelated scripts), that means upstream main has advanced independently on other work; this is normal and not part of the PR. Don't let it inflate the reported scope — cross-check with git show --stat <commit> for each commit in step 4 individually to get the true per-commit file list.

Phase 2: Write the report

Structure the report under these six headings. Base every claim on an actual diff read (git show <commit>, git diff <base>..HEAD -- <path>), not on commit message text alone — commit messages describe intent, not necessarily what the code does.

  1. Major purpose — 3-6 sentences synthesizing what the set of commits accomplishes and why, grouped by theme if the commits span multiple concerns.

  2. API interface changes — enumerate concretely:

    • GraphQL: new/changed types, fields, args in schema.graphql and the resolver file that backs them (pkg/admin/graphql/, pkg/portal/graphql/)
    • Go: new/changed exported struct fields, function signatures in pkg/lib/
    • HTTP/config: new endpoints, changed request/response shapes, authgear.yaml schema changes
    • Note whether each change is additive (safe) or a rename/removal (breaking)

Read the full file on GitHub · 101 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. 3d ago Changed · +9 lines 46937f004e39
  2. 5d ago First seen · 92 lines · 72 tokens per session scan A 9fd8c60f1998

Subscribe to this mod's changes

review-pr is a skill published in the GitHub repository authgear/authgear-server (2,024 stars, last pushed today), licensed Apache-2.0. It adds 72 tokens to every session and 4,502 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.