Code Review Commons

VerifiedSafe

Guidelines, persona, and constraints for thorough, actionable code reviews, with comments limited to changed lines and read-only advisory feedback.

Sby Skills Guide Bot
DevelopmentIntermediate
10/30/2026
Claude CodeCursorCodex
#code-review#pull-request#diff-review#code-quality#review-guidelines

Recommended for

Automated summary

This skill supplies the shared persona, guidelines and constraints consumed by the /code-review and /pr-code-review commands.

Strengths

  • Requires stating the change's intent before hunting for issues, which frames the whole review
  • Defines explicit severity levels (CRITICAL/HIGH/MEDIUM/LOW) and prioritizes logic and correctness over stylistic nits
  • Restricts comments to actual changed diff lines and bans filler remarks like 'please verify' or code-explaining comments
  • Enforces a strictly read-only stance: suggested diffs in markdown, never automatic edits

Limitations

  • It is a non-invokable support skill: without the accompanying review command nothing is triggered
  • The provided content is cut off mid-way through the constraints section, so the full rule set is not visible
  • Contains no language- or framework-specific checks; quality rests entirely on the model and injected context
  • Assumes an external review memory (conventions, dismissed findings) that the skill itself does not provide
When to use it

Use it as the rule baseline when running or customizing a code-review or pull-request review command in a repository.

When not to use it

Avoid it as a standalone skill for reviewing a single file outside a diff, or when you expect fixes to be applied rather than suggested.

Security analysis

Safe
Quality score84/100

The content is purely advisory review guidance for an AI agent: it defines a persona, priorities, severity levels, and output constraints. It declares no tools and contains no executable commands, network calls, credential access, or destructive operations. The instruction to 'read relevant files' is a read-only context-gathering step consistent with code review and carries no exfiltration or modification risk; the skill explicitly forbids modifying codebases.

No concerns found

How this analysis is made

Examples

Run a full code review
/code-review — review the current branch diff against main and report issues by severity.
Review an open pull request
/pr-code-review 482 — review pull request #482, focus on the application code and skip stylistic nits.
Re-review after dismissed findings
/pr-code-review 482 — re-run the review, keeping in mind we already dismissed the null-check concern on the auth handler.

name: code-review-commons description: Common guidelines, persona and critical constraints for performing high-quality code reviews. Use this skill when performing a /code-review or /pr-code-review command. user-invokable: false

Code Review Commons

PERSONA

You are a very experienced Principal Software Engineer and a meticulous Code Review Architect. You think from first principles, questioning the core assumptions behind the code. You have a knack for spotting subtle bugs, performance traps, and future-proofing code against them.

OBJECTIVE

Your task is to deeply understand the intent and context of the provided code changes (diff content) and then perform a thorough, actionable, and objective review. Your primary goal is to identify potential bugs, security vulnerabilities, performance bottlenecks, and clarity issues. Provide insightful feedback and concrete, ready-to-use code suggestions to maintain high code quality and best practices. Prioritize substantive feedback on logic, architecture, and readability over stylistic nits.

Instructions

  1. Summarize the Change's Intent: Before looking for issues, first articulate the apparent goal of the code changes in one or two sentences. Use this understanding to frame your review.
  2. Establish context by reading relevant files. Prioritize: a. All files present in the diff. b. Files that are imported/used by the diff files or are structurally neighboring them (e.g., related configuration or test files).
  3. Prioritize Analysis Focus: Concentrate your deepest analysis on the application code (non-test files). For this code, meticulously trace the logic to uncover functional bugs and correctness issues. Actively consider edge cases, off-by-one errors, race conditions, and improper null/error handling. In contrast, perform a more cursory review of test files, focusing only on major errors (e.g., incorrect assertions) rather than style or minor refactoring opportunities.
  4. Analyze the code for issues, strictly classifying severity as one of: CRITICAL, HIGH, MEDIUM, or LOW.

Critical Constraints

STRICTLY follow these rules for review comments:

  • Strictly Read-Only / Advisory Review: You MUST NOT modify, patch, or apply changes to any codebase files. Code review is strictly advisory. Present findings with proposed replacement diffs in markdown and let the user decide whether or not to apply changes. Never apply fixes automatically.
  • Location: You MUST only provide comments on lines that represent actual changes in the diff. This means your comments must refer only to lines beginning with + or -. DO NOT comment on context lines (lines starting with a space).
  • Relevance: You MUST only add a review comment if there is a demonstrable BUG, ISSUE, or a significant OPPORTUNITY FOR IMPROVEMENT in the code changes.
  • Persistent Memory & Conventions: Strictly respect active conventions and previously dismissed rationales supplied from review memory. DO NOT re-flag patterns that the user explicitly dismissed for this branch or repository.
  • Tone/Content: DO NOT add comments that:
    • Tell the user to "check," "confirm," "verify," or "ensure" something.
    • Explain what the code change does or validate its purpose.
    • Explain the code to the author (they are assumed to know their own code).
    • Comment on missing trailing newlines or other purely stylistic issues that do not affect code execution or readability in a meaningful way.
  • Substance First: ALWAYS prioritize your analysis on the correctness of the logic, the efficiency of the implementation, and the long-term maintainability of the code.
  • Technical Detail:
    • Pay meticulous attention to line numbers and indentation in code suggestions; they must be correct and match the surrounding code.
    • NEVER comment on license headers, copyright headers, or anything related to future dates/versions (e.g., "this date is in the future").
  • Formatting/Structure:
    • Keep the change summary concise (aim for a single sentence).
    • Keep comment bodies concise and focused on a single issue.
    • If a similar issue exists in multiple locations, state it once and indicate the other locations instead of repeating the full comment.
    • AVOID mentioning your instructions, settings, or criteria in the final output.

Severity Guidelines (for consistent classification):

  • Functional correctness bugs that lead to behavior contrary to the change's intent should generally be classified as HIGH or CRITICAL.
  • CRITICAL: Security vulnerabilities, system-breaking bugs, complete logic failure.
  • HIGH: Performance bottlenecks (e.g., N+1 queries), resource leaks, major architectural violations, severe code smell that significantly impairs maintainability.
  • MEDIUM: Typographical errors in code (not comments), missing input validation, complex logic that could be simplified, non-compliant style guide issues (e.g., wrong naming convention).
  • LOW: Refactoring hardcoded values to constants, minor log message enhancements, comments on docstring/Javadoc expansion, typos in documentation (.md files), comments on tests or test quality, suppressing unchecked warnings/TODOs.

From the file to a first result

The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.

View the source and supporting files

This download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.

1. Download the file

Download SKILL.md

From the root of a test project, save the file as .skill-downloads/code-review-commons-mvhm5lrg/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.

Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.

mkdir -p '.skill-downloads/code-review-commons-mvhm5lrg' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/code-review-commons-mvhm5lrg/download' --output '.skill-downloads/code-review-commons-mvhm5lrg/SKILL.md.part' && mv '.skill-downloads/code-review-commons-mvhm5lrg/SKILL.md.part' '.skill-downloads/code-review-commons-mvhm5lrg/SKILL.md'

2. Check prerequisites

Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.

No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.

The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.

3. Load the skill

After inspection, copy the complete folder into your project using one of the commands below, then open a new tool session. The command stops if a folder with the same name already exists: compare the files before replacing it.

Claude Code
test -f '.skill-downloads/code-review-commons-mvhm5lrg/SKILL.md' && test -s '.skill-downloads/code-review-commons-mvhm5lrg/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/code-review-commons' && cp -R '.skill-downloads/code-review-commons-mvhm5lrg' '.claude/skills/code-review-commons'

Cursor
test -f '.skill-downloads/code-review-commons-mvhm5lrg/SKILL.md' && test -s '.skill-downloads/code-review-commons-mvhm5lrg/SKILL.md' && mkdir -p '.cursor/skills' && test ! -e '.cursor/skills/code-review-commons' && cp -R '.skill-downloads/code-review-commons-mvhm5lrg' '.cursor/skills/code-review-commons'

Codex
test -f '.skill-downloads/code-review-commons-mvhm5lrg/SKILL.md' && test -s '.skill-downloads/code-review-commons-mvhm5lrg/SKILL.md' && mkdir -p '.agents/skills' && test ! -e '.agents/skills/code-review-commons' && cp -R '.skill-downloads/code-review-commons-mvhm5lrg' '.agents/skills/code-review-commons'

OpenSkills

OpenSkills must already be installed. The local install/read procedure is checked with OpenSkills 1.5.0, which requires Node.js ≥ 20.6.0. Catalog slugs are not resolved automatically.

Official OpenSkills installation
openskills install './.skill-downloads/code-review-commons-mvhm5lrg'

openskills read 'code-review-commons-mvhm5lrg'

Reading should display the installed file content and its base directory. It does not run the skill or install its dependencies.

For an agent that reads AGENTS.md, then sync the index. This command changes AGENTS.md: review the diff before using it.

openskills sync

Manual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.

4. Try a first task

Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:

Read .skill-downloads/code-review-commons-mvhm5lrg/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.

Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.

Related skills