New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

PR Review

by wysaid in wysaid/android-gpuimage-plus

Address review comments and CI failures for the current branch's PR

MITAuto-check passedDevelopment

Install PR Review

skills CLI
$ npx skills add wysaid/android-gpuimage-plus --skill pr-review -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install wysaid/android-gpuimage-plus pr-review --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/wysaid/android-gpuimage-plus.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.github/skills/pr-review .claude/skills/pr-review && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
pr-review
GitHub stars
1.9k
Token cost
~1.4k tokens
SKILL.md length
584 words
Files
1
Skills in repo
2
Repo updated
First seen
Licence
MIT

At a glance

Address review comments and CI failures for the current branch's PR

  • Works in 2 steps: Best practice? — Does the change follow… → Safe & trivial? — Is the diff mechanical…
  • Tasks that involve Pull requests
  • SKILL.md covers Constraints, Procedure and Output
  • Calls gh and python3

What it does

PR Review is an agent skill from wysaid/android-gpuimage-plus. Address review comments and CI failures for the current branch's PR

Its SKILL.md is about 1.4k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Development, covering Pull requests and Failing and flaky tests. It works with Android. The repository describes itself as: Android Image & Camera Filters Based on OpenGL. The licence is MIT.

When your agent uses it

  • Tasks that involve Pull requests
  • Tasks that involve Failing and flaky tests

Example prompts

  • “/pr-review”

Requirements

  • Python 3

Workflow steps

2 steps, taken from the first numbered list in SKILL.md.

  1. Best practice? — Does the change follow language/framework conventions (e.g. prefer imports over FQNs, use existing imports, standard…
  2. Safe & trivial? — Is the diff mechanical with zero risk of behavioral change and low effort?

What it can do on your machine

Read from SKILL.md and the folder at commit 32bf703. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • gh
    • python3

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use gh, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

PR Review loads about 1.4k tokens when it runs. Until then it costs about 19 tokens; SKILL.md has 584 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~19
When it runs · the whole SKILL.md, loaded when a task matches
~1.4k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from wysaid/android-gpuimage-plus at commit 32bf703, republished under its MIT licence (© wysaid). 584 words, ~1,413 tokens.

Download SKILL.mdSave it as .claude/skills/pr-review/SKILL.md (or your agent's skills folder).
name
pr-review
description
Address review comments and CI failures for the current branch's PR

Constraints

  • Prepend GH_PAGER= to every gh command (bash/zsh), or set $env:GH_PAGER="" in PowerShell — never modify global config
  • Key commands: gh pr list --head <BRANCH> · gh pr view <PR> --comments · gh pr checks <PR> · gh run view <RUN_ID> --log-failed
  • Workflow fixes cannot be verified locally — the fix is only confirmed once the remote CI re-runs and passes
Resolving Review Threads

After fixing a P1/P2 comment, mark its thread as resolved on GitHub using the GraphQL mutation:

bash
GH_PAGER= gh api graphql -f query='mutation { resolveReviewThread(input: { threadId: "PRRT_xxxx" }) { thread { isResolved } } }'

To get thread IDs with their status and a brief description:

bash
GH_PAGER= gh api graphql -f query='
{
  repository(owner: "OWNER", name: "REPO") {
    pullRequest(number: PR_NUMBER) {
      reviewThreads(first: 50) {
        nodes {
          id
          isResolved
          isOutdated
          path
          line
          comments(first: 1) { nodes { body } }
        }
      }
    }
  }
}' | python3 -c "
import json,sys
data = json.load(sys.stdin)
for t in data['data']['repository']['pullRequest']['reviewThreads']['nodes']:
    body = t['comments']['nodes'][0]['body'][:100].replace('\n',' ') if t['comments']['nodes'] else ''
    print(f\"ID={t['id']}, resolved={t['isResolved']}, outdated={t['isOutdated']}, {t['path']}:{t['line']}\")
    print(f\"  >> {body}\")
"

Also resolve threads that are outdated (the underlying code they referenced has since changed), as they are no longer actionable.

Review Restraint Policy

Before acting on any review comment or suggestion, classify it by importance:

LevelCriteriaAction
P1 — Must FixBug, security issue, broken behavior, API contract violation, CI failureFix immediately
P2 — Should FixCorrectness risk, meaningful maintainability improvement, clear code smell with real impactFix with brief justification
P3 — ConditionalStyle preference, minor naming, "could be cleaner"Fix if it aligns with best practices and the change is safe + trivial; defer if complex or has any potential functional impact
P4 — RejectContradicts project conventions, introduces unnecessary complexity, or is factually wrongReject with explanation

Rules:

  • For P3, apply this two-question test before touching the code:
    1. Best practice? — Does the change follow language/framework conventions (e.g. prefer imports over FQNs, use existing imports, standard patterns)?
    2. Safe & trivial? — Is the diff mechanical with zero risk of behavioral change and low effort?
    • Both yes → fix it silently, mark thread resolved.
    • Either no → do NOT modify code; record in the summary table and let the user decide.
  • Do not add comments to code unless the comment explains non-obvious logic that is truly necessary. Never add comments just to acknowledge a review suggestion was applied.
  • When the two-question test is ambiguous, prefer deferring rather than guessing.
Show full SKILL.md (273 more words)Show less

Procedure

  1. Locate PR — get PR number for current branch
  2. Fix CI failures — for each failing check:
    • Fetch logs: gh run view <RUN_ID> --log-failed
    • Workflow issue (wrong config, missing step, bad path):
      1. Fix .github/workflows/ directly
      2. Commit & push the workflow change
      3. Wait for the re-triggered run to complete: poll with GH_PAGER= gh pr checks <PR> (or GH_PAGER= gh run watch <RUN_ID>) until the affected check finishes
      4. If it passes → continue to the next failing check
      5. If it still fails → fetch new logs (gh run view <NEW_RUN_ID> --log-failed) and repeat from step i
    • Code issue, non-breaking: fix source code directly
    • Code issue, breaking change required: stop and report to developer for a decision
  3. Address review comments — apply the Review Restraint Policy to each comment:
    • P1/P2: implement the fix, then immediately resolve the thread using the GraphQL mutation above
    • Outdated threads: resolve them regardless of priority (no action needed, just mark resolved)
    • P3: apply the two-question test — fix + resolve if both answers are yes; otherwise record in summary and defer to user
    • P4: record rejection reason in summary
  4. Commit & push — single commit covering all non-workflow fixes (workflow fixes are pushed incrementally during step 2)
  5. Final verification — once all fixes are applied, confirm every check is green: GH_PAGER= gh pr checks <PR>

Output

  • Per CI failure: root cause and resolution (or escalation reason)
  • After all fixes: summary of check statuses confirming all green
  • Review summary table — produced at the end of every review session:
markdown
| # | Source (comment / CI) | Issue description | Priority | Action taken | Reason if not fixed |
|---|-----------------------|-------------------|----------|--------------|---------------------|
| 1 | Reviewer @xxx         | use import instead of FQN | P3  | Fixed        | Best practice + trivial mechanical change |
| 2 | Reviewer @xxx         | rename internal var foo→bar | P3 | Not fixed  | Non-standard opinion, no best-practice backing — deferred to user |
| 3 | CI: lint              | null-check missing | P1      | Fixed        | —                   |

All P3/P4 items that were not fixed must appear in this table with a clear reason, so the user can make an informed decision.

© wysaid, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .github/skills/pr-review of wysaid/android-gpuimage-plus.

Open the folder on GitHubat commit 32bf703

Used in 1 other repository

We found 1 copy of this SKILL.md (exact, near-identical or edited) in other folders. This page covers the copy in wysaid/android-gpuimage-plus, which our catalogue first saw on October 7, 2026.

Compare with similar skills

PR Review next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

PR Review compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
PR Review this skillwysaid/android-gpuimage-plus1.9k—~1.4kAutomated safety check: PassMIT
Stewardyschimke/compose-ai-tools117—~2kAutomated safety check: PassApache-2.0
PR Babysitteropeninterpreter/openinterpreter69k3 repos~4.2kAutomated safety check: PassApache-2.0
Iterate PRmeshery/meshery-operator1517 repos~2.2kAutomated safety check: PassApache-2.0
PlotJuggler Ship CheckPlotJuggler/PlotJuggler6.2k—~1.3kAutomated safety check: PassMPL-2.0
GitHub PR Imagesbikeindex/bike_index308—~1.9kAutomated safety check: PassAGPL-3.0

Similar skills

  • Steward

    yschimke/compose-ai-tools

    Drive a pull request on this repository to green — which fast checks to run before pushing, how to read a red check, and what to do about a review comment.

    117 GitHub stars~2k tokensUpdated today
    DevelopmentAuto-check passed

  • PR Babysitter

    openinterpreter/openinterpreter

    Watches an open GitHub pull request until it merges, handling review comments, diagnosing CI failures and retrying flaky checks along the way.

    69k GitHub starsUsed in 3 repos~4.2k tokens
    DevelopmentAuto-check passed

  • Iterate PR

    meshery/meshery-operator

    Iterate on a PR until CI passes. An agent skill from meshery/meshery-operator.

    151 GitHub starsUsed in 7 repos~2.2k tokens
    DevelopmentAuto-check passed

  • PlotJuggler Ship Check

    PlotJuggler/PlotJuggler

    Runs a gated finish-line checklist before committing a PlotJuggler PJ4 change: build proof, red-test triage, hooks, docs freshness and a diff self-review.

    6.2k GitHub stars~1.3k tokensUpdated 8 days ago
    DevelopmentAuto-check passed

  • GitHub PR Images

    bikeindex/bike_index

    Embed a local image file into an existing GitHub PR — either in the PR body or as a comment.

    308 GitHub stars~1.9k tokensUpdated today
    DevelopmentAuto-check passed

  • PR Draft Summary

    kernitus/BukkitOldCombatMechanics

    A skill your agent uses when drafting pull-request titles, descriptions, change summaries, risk notes, validation sections, or reviewer handoff text; do not use for implementation design, release…

    225 GitHub stars~526 tokensUpdated 5 days ago
    DevelopmentAuto-check passed

More from wysaid/android-gpuimage-plus

  • PR Submit

    wysaid/android-gpuimage-plus

    Create or update a GitHub Pull Request — use ONLY when explicitly asked to create/submit/open a PR

    1.9k GitHub stars~250 tokensUpdated 2 mo ago
    Auto-check passed

Works with

Categories

Questions about PR Review

What does PR Review do?

Address review comments and CI failures for the current branch's PR. PR Review is an agent skill from wysaid/android-gpuimage-plus.

When should I use PR Review?

PR Review fits situations like: tasks that involve Pull requests; tasks that involve Failing and flaky tests.

How do I install PR Review in Claude Code?

Run `npx skills add wysaid/android-gpuimage-plus --skill pr-review -a claude-code`. Or copy the skill folder (.github/skills/pr-review in wysaid/android-gpuimage-plus) into .claude/skills/pr-review in your project. Claude Code loads it when a task matches its description.

How do I install PR Review in Codex?

Run `npx skills add wysaid/android-gpuimage-plus --skill pr-review -a codex`. Or copy the skill folder (.github/skills/pr-review in wysaid/android-gpuimage-plus) into .agents/skills/pr-review in your project. Codex loads it when a task matches its description.

Can I use PR Review in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add wysaid/android-gpuimage-plus --skill pr-review -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/pr-review, .gemini/skills/pr-review, .github/skills/pr-review and .opencode/skills/pr-review in your project.

What does PR Review need to run?

Going by SKILL.md and its folder, PR Review needs the command-line tools its instructions call (gh and python3). Our summary lists: Python 3.

Does PR Review access the network?

SKILL.md contains no URLs. Its commands use gh, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is PR Review safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does PR Review use?

PR Review is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does PR Review use?

About 1.4k tokens (SKILL.md is roughly 5.7k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to PR Review?

Skills that share tags, products or a category with PR Review: Steward (yschimke/compose-ai-tools, 117 stars), PR Babysitter (openinterpreter/openinterpreter, 69k stars), Iterate PR (meshery/meshery-operator, 151 stars) and PlotJuggler Ship Check (PlotJuggler/PlotJuggler, 6.2k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains PR Review?

wysaid (a GitHub user) maintains it in wysaid/android-gpuimage-plus, which has 1,934 GitHub stars. The repository holds 2 skills in this directory. The repository was last updated on July 28, 2026.

Source: wysaid/android-gpuimage-plus on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.