MCPInstall

code-review

Review code changes for correctness, regressions, and unintended scope. Use when reviewing a pull request, branch, commit, patch, or local diff, including when invoked with /code-review.

Installs
60
Last 24h
—
Trending rank
—
Security
Not audited

Install

Works with every agent the skills CLI supports.

npx skills add https://github.com/microsoft/vscode --skill code-review
  1. 01

    Install

    Run the command in your project. The CLI asks which agents to install for.

  2. 02

    Describe the task

    Ask your agent for something this skill covers. It loads code-review automatically when relevant.

  3. 03

    Review the result

    The agent follows the SKILL.md steps, scripts and references bundled with the skill.

SKILL.md

The instructions your agent reads when the skill is triggered.

Code Review

Review the proposed changes as a senior maintainer. Focus on actionable defects introduced by the changes, especially correctness issues, regressions, and changes that are not necessary for the stated goal.

Establish the Review Scope

Before reviewing:

  1. Determine the intended change from the request, pull request description, linked issue, and maintainer comments.
  2. Identify the complete diff against the appropriate base.
  3. Account for every changed file and subsystem.

Do not infer the purpose solely from the implementation. If the intended behavior is unclear, state the assumption used for the review.

Review the Changes

Trace the changed behavior through its callers, dependencies, and tests. Check for:

  • Incorrect logic, broken edge cases, or behavior that contradicts the stated goal.
  • Regressions in existing behavior, including platform-specific and lifecycle behavior.
  • Missing or incorrect error handling.
  • Unsafe assumptions about data, ordering, concurrency, or object lifetime.
  • Tests that do not cover the changed behavior or that would pass without the fix.
  • Changes whose connection to the stated goal is unclear.

Keep fixes tied to the underlying cause. Do not accept unrelated changes to behavior, state management, or architecture merely because they make the reported symptom disappear.

Treat prerequisite or tightly coupled changes as in scope only when their necessity is clear from the implementation or pull request description. A changed file should be necessary for the stated goal or be a directly related test, documentation update, generated artifact, or required refactor. Independent cleanup and speculative improvements belong in a separate change.

Report Findings

Report only findings that are:

  • Introduced or made materially worse by the changes under review.
  • Specific and actionable.
  • Supported by the code or by a concrete missing validation.

For each finding:

  1. State the user-visible or engineering impact.
  2. Explain the conditions that trigger it.
  3. Point to the smallest relevant changed line or range.
  4. Explain why the current implementation is incorrect.

Prioritize findings by severity. Do not report stylistic preferences, speculative concerns, or pre-existing problems unrelated to the change. If no actionable findings remain, say so explicitly and mention any meaningful validation gaps.

Flag changes whose connection to the stated goal is unclear even when tests pass or the resulting behavior appears correct. Ask for the causal link to be explained; if none exists, recommend reverting the change or moving it to a separate pull request.

FAQ

What does the code-review skill do?

Review code changes for correctness, regressions, and unintended scope. Use when reviewing a pull request, branch, commit, patch, or local diff, including when invoked with /code-review.

How do I install code-review?

Run npx skills add https://github.com/microsoft/vscode --skill code-review in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/code-review/ (personal) or .claude/skills/code-review/ (project).

Which AI agents can use code-review?

Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.

Is code-review safe to install?

No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.

Who maintains code-review?

It is published by microsoft in microsoft/vscode. It has 60 recorded installs.

More from microsoft/vscode

View all →

memory-leak-audit

microsoft/vscode

Audit code for memory leaks and disposable issues. Use when reviewing event listeners, DOM handlers, lifecycle callbacks, or fixing leak reports. Covers addDisposableListener, Event.once, MutableDisposable, DisposableStore, and onWillDispose patterns.

OfficialSecurity3.2K installs

azure-pipelines

microsoft/vscode

Use when validating Azure DevOps pipeline changes for the VS Code build. Covers queueing builds, checking build status, viewing logs, and iterating on pipeline YAML changes without waiting for full CI runs.

OfficialDevOps & Cloud3.2K installs

accessibility

microsoft/vscode

Primary accessibility skill for VS Code. REQUIRED for new feature and contribution work, and also applies to updates of existing UI. Covers accessibility help dialogs, accessible views, verbosity settings, signals, ARIA announcements, keyboard navigation, and ARIA labels/roles.

OfficialFrontend & Design3.2K installs

sessions

microsoft/vscode

Core principles and workflow router for changes to the Agents Window under src/vs/sessions.

OfficialProductivity & Automation3.1K installs

author-contributions

microsoft/vscode

Identify all files a specific author contributed to on a branch vs its upstream, tracing code through renames. Use when asked who edited what, what code an author contributed, or to audit authorship before a merge. This skill should be run as a subagent — it performs many git operations and returns a concise table.

OfficialAI & Agents3K installs

update-screenshots

microsoft/vscode

Update the committed blocks-ci screenshot hashes after the "Screenshots & Tests" check fails, or investigate a screenshot diff reported on a PR. Use when asked to update, accept, or refresh component screenshot baselines from CI. This skill should be run as a subagent.

OfficialDevOps & Cloud3K installs

Popular code quality & workflow skills

View category →

grill-me

mattpocock/skills

A relentless interview to sharpen a plan or design.

Scan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.

handoff

mattpocock/skills

Compact the current conversation into a handoff document for another agent to pick up.

domain-modeling

mattpocock/skills

Build and sharpen a project's domain model. Use when discussing codebase terminology, writing or editing a GLOSSARY.md, or recording or editing an ADR.

codebase-design

mattpocock/skills

Shared vocabulary for designing deep modules. Use when the user wants to design or improve a module's interface, find deepening opportunities, decide where a seam goes, make code more testable or AI-navigable, or when another skill needs the deep-module vocabulary.

diagnosing-bugs

mattpocock/skills

Diagnosis loop for hard bugs and performance regressions. Use when the user says "diagnose"/"debug this", or reports something broken/throwing/failing/slow.