Agent skill

Review PR

by razorpay in razorpay/blade

Review blade PRs by fetching diff, checking CI status, and getting Storybook URL.

MITAuto-check passedDevelopment

Install Review PR

skills CLI
$ npx skills add razorpay/blade --skill review-pr -a claude-code

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

GitHub CLI
$ gh skill install razorpay/blade review-pr --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/razorpay/blade.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/review-pr .claude/skills/review-pr && 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
review-pr
GitHub stars
656
Token cost
~1.1k tokens
SKILL.md length
358 words
Files
3 (incl. scripts, references)
Skills in repo
16
Repo updated
First seen
Licence
MIT

At a glance

Review blade PRs by fetching diff, checking CI status, and getting Storybook URL.

  • Works in 5 steps: Fetch the diff and PR metadata → Spawn Critique Subagents in parallel → Synthesize → …
  • Checking PR status
  • SKILL.md covers Arguments / Clarifications, Prerequisites, Instructions and Steps
  • Runs JavaScript scripts from its folder; calls gh, node and npm

What it does

Review PR is an agent skill from razorpay/blade. Review blade PRs by fetching diff, checking CI status, and getting Storybook URL. Use when reviewing PRs, checking PR status, or getting preview links for razorpay/blade PRs.

Its SKILL.md is about 1.1k tokens, which your agent loads only when the skill is triggered. The skill folder holds 4 other files, including scripts and reference files (for example `references/output-format.md` and `scripts/post-review.js`).

It sits in Development, covering Pull requests. It works with Storybook. The repository describes itself as: Design System that powers Razorpay. The licence is MIT.

When your agent uses it

  • Checking PR status
  • Getting preview links for razorpay/blade PRs

Example prompts

  • “/review-pr”

Requirements

  • Node.js

Workflow steps

5 steps, taken from the step headings in SKILL.md.

  1. Fetch the diff and PR metadata
  2. Spawn Critique Subagents in parallel
  3. Synthesize
  4. Spawn filter-critique subagent
  5. Post review to GitHub

What it can do on your machine

Read from SKILL.md and the folder at commit fa38cf0. 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

    Ships 1 file in scripts/ (JavaScript), which the agent can run.

    Shell commands in SKILL.md call:

    • gh
    • node
    • npm

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

  • Network

    No URLs in SKILL.md. Its commands use gh and npm, 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

Review PR loads about 1.1k tokens when it runs, and up to ~1.9k if it reads all its reference files. Until then it costs about 46 tokens; SKILL.md has 358 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~46
When it runs · the whole SKILL.md, loaded when a task matches
~1.1k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~1.9k

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); the scripts in this folder are not scanned.

SKILL.md

The full file from razorpay/blade at commit fa38cf0, republished under its MIT licence (© razorpay). 358 words, ~1,119 tokens.

Download SKILL.mdSave it as .claude/skills/review-pr/SKILL.md (or your agent's skills folder). This skill also uses 2 other files; get the full folder from GitHub.
name
review-pr
description
Review blade PRs by fetching diff, checking CI status, and getting Storybook URL. Use when reviewing PRs, checking PR status, or getting preview links for razorpay/blade PRs.
disable-model-invocation
true

Review Blade PR

Arguments / Clarifications

  • PR: Link / Number of the PR to review
  • ShouldReviewUI: Whether UI should be reviewed or not. Clarify if not mentioned.
  • ShouldRunHeadedBrowser: Whether to run the browser in headed mode. Clarify if not mentioned.
  • ShouldSubmitReview: Whether to submit the review as a live comment or keep it as a pending draft. Default: true. If false, the review is saved as pending and not visible on the PR until submitted manually.

Prerequisites

  • GitHub CLI (gh) must be installed and authenticated
  • agent-browser command must be installed (if ShouldReviewUI is true). If not installed, go ahead and install it.
    sh
    npm install -g agent-browser
    agent-browser install

Instructions

  • You should never add any review.md file in PR or anywhere in the codebase. post-review script takes care of posting review and if it fails, you can add review as a comment on the PR
  • If you're unable to spawn the subagents, stop the skill immediately and respond with error message.
    json
    {
      "overview-comment": {},
      "inlined-comments": [
        {
          "file": "",
          "line": 0,
          "side": "RIGHT",
          "severity": "critical",
          "confidence": 10,
          "critique": "code-quality-critique",
          "problem": "Could not spawn the subagents. Please check the logs for more details."
        }
      ]
    }

Steps

1. Fetch the diff and PR metadata

Run these in parallel:

bash
gh api repos/razorpay/blade/pulls/{PR_NUMBER}/files --jq '[.[] | select(.filename | test("lock$|lock\\.json$|lock\\.yaml$") | not) | {filename, patch}]'
bash
gh pr view {PR_NUMBER} --repo razorpay/blade --json title,body

Store the outputs as DIFF, PR_TITLE, and PR_BODY.

2. Spawn Critique Subagents in parallel

code-quality-critique:

subagent_type: code-quality-critique
prompt: |
  PR_NUMBER={PR_NUMBER}
  PR_TITLE={PR_TITLE}
  PR_BODY={PR_BODY}
  DIFF={DIFF}

api-decision-critique (only if the diff contains component changes):

subagent_type: api-decision-critique
prompt: |
  PR_NUMBER={PR_NUMBER}
  PR_TITLE={PR_TITLE}
  PR_BODY={PR_BODY}
  DIFF={DIFF}

ui-critique (only if ShouldReviewUI is true):

subagent_type: ui-critique
prompt: |
  PR_NUMBER={PR_NUMBER}
  HEADED={ShouldRunHeadedBrowser}
  PR_TITLE={PR_TITLE}
  PR_BODY={PR_BODY}
  DIFF={DIFF}
Show full SKILL.md (165 more words)Show less
3. Synthesize

Combine all subagent outputs into a single JSON object. Follow the output format in references/output-format.md exactly.

4. Spawn filter-critique subagent
subagent_type: filter-critique
prompt: |
  PR_NUMBER={PR_NUMBER}
  INLINED_COMMENTS={INLINED_COMMENTS}

Replace inlined-comments in the review JSON with the filtered array returned by the agent.

Then recalculate reviewStatus based on the filtered array:

  • Set reviewStatus to 'approved' if:
    • Check reviewStatusReason to see if reviewStatus was not approved only because of inlined comments array before filtering and that changes now. If so, set reviewStatus to 'approved' now since the issues are filtered out now.
  • Otherwise, set reviewStatus to 'commented'

Update reviewStatus in the review JSON before proceeding to Step 5.

5. Post review to GitHub

Save the synthesized JSON to a temp file and run the post script. Pass --pending if ShouldSubmitReview is false — in that case the review is saved as a draft and will not be visible on the PR until submitted manually.

bash
# Save the review JSON to a temp file
cat > /tmp/blade-review-{PR_NUMBER}.json << 'EOF'
{REVIEW_JSON}
EOF

# If ShouldPostComment is true (default):
node .agents/skills/review-pr/scripts/post-review.js /tmp/blade-review-{PR_NUMBER}.json {PR_NUMBER}

# If ShouldPostComment is false:
node .agents/skills/review-pr/scripts/post-review.js /tmp/blade-review-{PR_NUMBER}.json {PR_NUMBER} razorpay/blade --pending

When --pending is used, the script prints the review ID and a command to submit it when ready:

bash
# To submit the pending review later:
gh api repos/razorpay/blade/pulls/{PR_NUMBER}/reviews/{REVIEW_ID}/events --method POST --field event=COMMENT

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

Files

SKILL.md and 2 other files (scripts, references) in .agents/skills/review-pr of razorpay/blade.

  • SKILL.md
  • references/output-format.md
  • scripts/post-review.js

Open the folder on GitHubat commit fa38cf0

Compare with similar skills

Review PR 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.

Review PR compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Review PR this skillrazorpay/blade656—~1.1kAutomated safety check: PassMIT
Handoffclosedloop-ai/claude-plugins122—~5kAutomated safety check: PassApache-2.0
Astro PR Writerwithastro/astro63k—~1.2kAutomated safety check: PassCustom licence
Docs AuthoringTracecatHQ/tracecat3.8k—~3.1kAutomated safety check: NotesAGPL-3.0
Clean Vmaxeldelafosse/loop281—~906Automated safety check: PassMIT
Create PRUniClipboard/UniClipboard1.9k—~4.6kAutomated safety check: PassAGPL-3.0

Similar skills

  • Handoff

    closedloop-ai/claude-plugins

    Finish a vibe session in symphony-alpha and hand it to whoever picks it up next, usually design and then engineering.

    122 GitHub stars~5k tokensUpdated yesterday
    DevelopmentAuto-check passed
  • Astro PR Writer

    withastro/astro

    Official

    Write and update Astro pull requests with reviewer-friendly titles and high-signal bodies.

    63k GitHub stars~1.2k tokensUpdated today
    DevelopmentAuto-check passed
  • Docs Authoring

    TracecatHQ/tracecat

    A skill your agent uses when adding or updating documentation pages in an existing docs site.

    3.8k GitHub stars~3.1k tokensUpdated yesterday
    DevelopmentAuto-check: notes
  • Clean Vm

    axeldelafosse/loop

    Safely clean the local loop VM by reporting and removing stale loop runs, inactive Next.js or Storybook servers, optional browser windows, and unused loop-created worktrees without disturbing active…

    281 GitHub stars~906 tokensUpdated 6 mo ago
    DevelopmentAuto-check passed
  • Create PR

    UniClipboard/UniClipboard

    Push the current branch and open a GitHub pull request against main.

    1.9k GitHub stars~4.6k tokensUpdated today
    DevelopmentAuto-check passed
  • Phoenix Storybook

    Arize-ai/phoenix

    Conventions for creating, modifying, and reviewing production-faithful Storybook stories in the Phoenix frontend (js/app/stories, js/app/.storybook).

    12k GitHub stars~1.9k tokensUpdated today
    Frontend & DesignAuto-check passed

More from razorpay/blade

All 16 skills in this repo
  • Migrate To Rn

    razorpay/blade

    Add React Native support (.native.tsx files) to Blade components that currently only have web implementations.

    656 GitHub stars~4.4k tokensUpdated today
    Auto-check passed
  • Migrate To Svelte

    razorpay/blade

    Orchestrate parallel migration of Blade React components to Svelte 5.

    656 GitHub stars~4.7k tokensUpdated today
    Auto-check passed
  • UI Code Guidelines

    razorpay/blade

    Important guidelines for writing frontend UI code. An agent skill from razorpay/blade.

    656 GitHub stars~866 tokensUpdated today
    Auto-check passed
  • Heal PR

    razorpay/blade

    Heal a Blade PR by fixing CI failures, missing changesets, and sanity issues.

    656 GitHub stars~414 tokensUpdated today
    Auto-check passed
  • This rule helps in reviewing the component before shipping to make sure all important things are checked

    656 GitHub stars~1.3k tokensUpdated today
    Auto-check passed
  • Resolve Comments

    razorpay/blade

    Resolve comments on a pull request. An agent skill from razorpay/blade.

    656 GitHub stars~720 tokensUpdated today
    Auto-check passed

Works with

Questions about Review PR

What does Review PR do?

Review blade PRs by fetching diff, checking CI status, and getting Storybook URL. Review PR is an agent skill from razorpay/blade. Review blade PRs by fetching diff, checking CI status, and getting Storybook URL.

When should I use Review PR?

Review PR fits situations like: checking PR status; getting preview links for razorpay/blade PRs.

How do I install Review PR in Claude Code?

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

How do I install Review PR in Codex?

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

Can I use Review PR 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 razorpay/blade --skill review-pr -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/review-pr, .gemini/skills/review-pr, .github/skills/review-pr and .opencode/skills/review-pr in your project.

What does Review PR need to run?

Going by SKILL.md and its folder, Review PR needs JavaScript for the scripts in its folder and the command-line tools its instructions call (gh, node and npm). Our summary lists: Node.js.

Does Review PR access the network?

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

Is Review PR 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. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Review PR use?

Review PR 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 Review PR use?

About 1.1k tokens (SKILL.md is roughly 4.5k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 825 tokens, read only when the agent opens those files.

What are the alternatives to Review PR?

Skills that share tags, products or a category with Review PR: Handoff (closedloop-ai/claude-plugins, 122 stars), Astro PR Writer (withastro/astro, 63k stars), Docs Authoring (TracecatHQ/tracecat, 3.8k stars) and Clean Vm (axeldelafosse/loop, 281 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Review PR?

razorpay (a GitHub organization) maintains it in razorpay/blade, which has 656 GitHub stars. The repository holds 16 skills in this directory. The repository was last updated on October 9, 2026.

Source: razorpay/blade on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.