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

Official agent skill

PR Review

by NVIDIA in NVIDIA/Megatron-LM

Review rubric for the /review pull-request command. An agent skill from NVIDIA/Megatron-LM.

OfficialApache-2.0Auto-check passedDevelopment

Install PR Review

skills CLI
$ npx skills add NVIDIA/Megatron-LM --skill pr-review -a claude-code

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

GitHub CLI
$ gh skill install NVIDIA/Megatron-LM 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/NVIDIA/Megatron-LM.git skills-src && mkdir -p .claude/skills && cp -r skills-src/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
18k
Token cost
~839 tokens
SKILL.md length
430 words
Files
3 (incl. references)
Skills in repo
14
Repo updated
First seen
Licence
Apache-2.0

At a glance

Review rubric for the /review pull-request command. An agent skill from NVIDIA/Megatron-LM.

  • Works in 6 steps: Read the PR diff first: gh pr diff… → From the changed files and areas,… → Read those SKILL.md files with the Read… → …
  • Tasks that involve Pull requests
  • SKILL.md covers Pick the depth, Mandatory workflow — never… and Posting findings
  • Calls gh

What it does

PR Review is an agent skill from NVIDIA/Megatron-LM, published by the product's own GitHub organization. Review rubric for the /review pull-request command. The formal reviewer reads it as a file and it is not an interactive skill — do not load it to answer questions or to review code outside that command.

Its SKILL.md is about 840 tokens, which your agent loads only when the skill is triggered. The skill folder holds 3 other files, including reference files (for example `references/light.md` and `references/strict.md`).

It sits in Development, covering Pull requests, Quizzes and assessments and Code review. It works with NVIDIA AI Platform. The repository describes itself as: Ongoing research training transformer models at scale. The licence is Apache-2.0.

When your agent uses it

  • Tasks that involve Pull requests
  • Tasks that involve Quizzes and assessments
  • Tasks that involve Code review

Example prompts

  • “/pr-review”

Requirements

  • Python 3

Workflow steps

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

  1. Read the PR diff first: gh pr diff $PR_NUMBER --repo $REPO.
  2. From the changed files and areas, identify the relevant domain skills. Use
  3. Read those SKILL.md files with the Read tool.
  4. Read the depth reference from the table above.
  5. For Python changes, read the local style-guide.md
  6. Only then review.

What it can do on your machine

Read from SKILL.md and the folder at commit d5fbb65. 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

    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 839 tokens when it runs, and up to ~4k if it reads all its reference files. Until then it costs about 54 tokens; SKILL.md has 430 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~54
When it runs · the whole SKILL.md, loaded when a task matches
~839
With references · SKILL.md plus every file in references/, read only if the agent opens them
~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 NVIDIA/Megatron-LM at commit d5fbb65, republished under its Apache-2.0 licence (© NVIDIA). 430 words, ~839 tokens.

Download SKILL.mdSave it as .claude/skills/pr-review/SKILL.md (or your agent's skills folder). This skill also uses 2 other files; get the full folder from GitHub.
name
pr-review
description
Review rubric for the `/review` pull-request command. The formal reviewer reads it as a file and it is not an interactive skill — do not load it to answer questions or to review code outside that command.
license
Apache-2.0
disable-model-invocation
true
user_invocable
false

PR Review

This is the review rubric behind the /review pull-request command. The reviewer receives the requested REVIEW DEPTH (mode=light|strict), then reads this file.

It lives in skills/ so the rubric can be diffed, reviewed and evolved like code instead of being buried in YAML, but it is deliberately inert: the frontmatter carries disable-model-invocation: true, so Claude Code drops it from the advertised skill list and refuses to auto-invoke it. Reading it by path, which is exactly what the reviewer does, still works. Do not add a when_to_use: field — that is the trigger text that would make it activate on its own.

Pick the depth

Read only the reference for the depth the caller passed. Each one is a complete rubric, so loading the other adds nothing but noise:

REVIEW DEPTHComment triggerRead
light/reviewskills/pr-review/references/light.md
strict/review mode=strictskills/pr-review/references/strict.md

Mandatory workflow — never skip or reorder

  1. Read the PR diff first: gh pr diff $PR_NUMBER --repo $REPO.
  2. From the changed files and areas, identify the relevant domain skills. Use Glob on skills/*/SKILL.md to see what exists rather than assuming names — the set changes over time, and Megatron-LM domain guides carry an mcore- prefix (mcore-testing, mcore-cicd, mcore-build-and-dependency, mcore-linting-and-formatting, mcore-run-on-slurm, mcore-split-pr, mcore-onboard-gb200-1node-tests, …).
  3. Read those SKILL.md files with the Read tool.
  4. Read the depth reference from the table above.
  5. For Python changes, read the local style-guide.md with the Read tool. Do not load the entire Google Python Style Guide by default. Consult relevant sections as needed; read it in full when performing a comprehensive style audit.
  6. Only then review.

The order is what makes the review worth reading. A reviewer who forms an opinion before loading mcore-testing will invent a test convention that this repo does not use, and a confidently wrong review comment costs the author more time than no review at all.

Show full SKILL.md (122 more words)Show less

Posting findings

Use inline ```suggestion blocks only for simple, self-contained line replacements — typos, renames, single-line fixes. For structural changes that add, remove or reorganize blocks of code (a new function, an inserted YAML step, reordered logic), post a top-level PR comment with a fenced code block showing the proposed change instead. GitHub's suggestion blocks can only replace the exact lines they are anchored to, so an insertion or a multi-block restructuring applied via suggestion silently corrupts the author's file.

Findings that deeper analysis invalidates should be dropped entirely rather than hedged. A hedged comment transfers the work of disproving it to the author.

Completion — what to post at the end, and when to approve — is depth-specific and covered in the reference file.

© NVIDIA, Apache-2.0. 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 (references) in skills/pr-review of NVIDIA/Megatron-LM.

  • SKILL.md
  • references/light.md
  • references/strict.md

Open the folder on GitHubat commit d5fbb65

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 skillNVIDIA/Megatron-LM18k—~839Automated safety check: PassApache-2.0
Orca ReviewContinuum-AI-Corp/Orca-Code-Review173—~3.5kAutomated safety check: PassMIT
Review PRmicrosoft/vscode-containers141—~900Automated safety check: PassCustom licence
Code Review Rubricmakifbaysal/tasktrooper109—~2.5kAutomated safety check: PassApache-2.0
PR Babysitteropeninterpreter/openinterpreter69k3 repos~4.2kAutomated safety check: PassApache-2.0
Understand Diff AnalysisEgonex-AI/Understand-Anything86k1 repos~1.4kAutomated safety check: PassMIT

Similar skills

  • Orca Review

    Continuum-AI-Corp/Orca-Code-Review

    Review code changes yourself, locally, with OrcaCode Review's severity contract and merge gate — no GitHub Action, no OrcaRouter account, no API key.

    173 GitHub stars~3.5k tokensUpdated 17 days ago
    DevelopmentAuto-check passed

  • Review PR

    microsoft/vscode-containers

    Official

    Review a specific vscode-containers pull request on demand from the CLI (or any interactive agent), the way a Container Tools maintainer would.

    141 GitHub stars~900 tokensUpdated today
    DevelopmentAuto-check passed

  • Code Review Rubric

    makifbaysal/tasktrooper

    A skill your agent uses when a task is in codereview - how to read the diff, which findings block, and the verdict move

    109 GitHub stars~2.5k 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

  • Understand Diff Analysis

    Egonex-AI/Understand-Anything

    Reads your git changes or a pull request against a prebuilt knowledge graph of the project to explain what changed, which components are affected and what is risky.

    86k GitHub starsUsed in 1 repo~1.4k tokens
    DevelopmentAuto-check passed

  • WooCommerce Code Review

    woocommerce/woocommerce

    Reviews WooCommerce code changes against the project's standards, flagging backend PHP architecture, naming, documentation, data integrity and testing violations.

    11k GitHub starsUsed in 3 repos~1.1k tokens
    DevelopmentAuto-check passed

More from NVIDIA/Megatron-LM

All 14 skills in this repo

  • Megatron Core Testing Guide

    NVIDIA/Megatron-LM

    Official

    Guide to the Megatron-LM test system: layout, recipe YAML, running and adding unit and functional tests, golden values, marker filters and CI parity.

    18k GitHub starsUsed in 1 repo~2.1k tokens
    Auto-check passed

  • Update Megatron Golden Values

    NVIDIA/Megatron-LM

    Official

    Refreshes stored golden values from a GitHub Actions run, reports signed percentage changes per model, and writes a summary ready for a pull request description.

    18k GitHub stars~2.8k tokensUpdated today
    Auto-check passed

  • Megatron-LM Container and Dependency Setup

    NVIDIA/Megatron-LM

    Official

    Walks an agent through working inside the Megatron-LM CI container and changing dependencies with uv, so lock files resolve the same locally and in CI.

    18k GitHub stars~2.6k tokensUpdated today
    Auto-check passed

  • Megatron-LM Base Image Bump

    NVIDIA/Megatron-LM

    Official

    Moves Megatron-LM CI to a newer NVIDIA PyTorch base image, updating both the GitHub and GitLab pins together and handling the CI follow-up.

    18k GitHub stars~2.8k tokensUpdated today
    Auto-check passed

  • Megatron-LM CI/CD Guide

    NVIDIA/Megatron-LM

    Official

    Explains Megatron-LM's CI pipeline, PR scope labels, triggering the internal GitLab CI with a dry run first, and investigating CI failures.

    18k GitHub stars~1.8k tokensUpdated today
    Auto-check passed

  • Megatron-LM CI Failure Triage

    NVIDIA/Megatron-LM

    Official

    Investigates a failing GitHub Actions run or job for Megatron-LM, finds the root cause plus the PR and test author involved, and files a structured bug issue.

    18k GitHub stars~1.6k tokensUpdated today
    Auto-check passed

Works with

Categories

Questions about PR Review

What does PR Review do?

Review rubric for the /review pull-request command. An agent skill from NVIDIA/Megatron-LM. PR Review is an agent skill from NVIDIA/Megatron-LM, published by the product's own GitHub organization. Review rubric for the /review pull-request command.

When should I use PR Review?

PR Review fits situations like: tasks that involve Pull requests; tasks that involve Quizzes and assessments; tasks that involve Code review.

How do I install PR Review in Claude Code?

Run `npx skills add NVIDIA/Megatron-LM --skill pr-review -a claude-code`. Or copy the skill folder (skills/pr-review in NVIDIA/Megatron-LM) 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 NVIDIA/Megatron-LM --skill pr-review -a codex`. Or copy the skill folder (skills/pr-review in NVIDIA/Megatron-LM) 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 NVIDIA/Megatron-LM --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). 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 Apache-2.0 licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does PR Review use?

About 839 tokens (SKILL.md is roughly 3.4k 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 3.1k tokens, read only when the agent opens those files.

What are the alternatives to PR Review?

Skills that share tags, products or a category with PR Review: Orca Review (Continuum-AI-Corp/Orca-Code-Review, 173 stars), Review PR (microsoft/vscode-containers, 141 stars), Code Review Rubric (makifbaysal/tasktrooper, 109 stars) and PR Babysitter (openinterpreter/openinterpreter, 69k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains PR Review?

NVIDIA (a GitHub organization, an official publisher) maintains it in NVIDIA/Megatron-LM, which has 18,083 GitHub stars. The repository holds 14 skills in this directory. The repository was last updated on October 8, 2026.

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