Skip to content
Agentmart
Submit a tool

reviewer-agent

Code review agent for quality assessment, security scanning, standards compliance, and professional review reports.

Skill by Celia-veey

IndexedMIT
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/celia-veey-reviewer-agent

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Code review agent for quality assessment, security scanning, standards compliance, and professional review reports. TRIGGER when: code changes are ready for review, PR needs approval, quality check required. DO NOT TRIGGER when: code is not yet implemented or still in draft.

Use cases: security, agent-building

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/reviewer-agent

What's inside

Skill name
reviewer-agent
Description
Code review agent for quality assessment, security scanning, standards compliance, and professional review reports. TRIGGER when: code changes are ready for review, PR needs approval, quality check required. DO NOT TRIGGER when: code is not yet implemented or still in draft.
License
MIT
Metadata
version: 1.0.0 · category: code-review
Instructions
379 lines in SKILL.md · view on GitHub

Agent Skills spec issues

  • name: "reviewer-agent" does not match folder "Reviewer"
  • files: referenced but missing: references/api-design.md, references/environment-management.md, references/auth-flow.md, references/testing-strategy.md

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Reviewer Agent

## Role
You are a senior code review expert responsible for multi-dimensional code quality assessment to ensure code is standardized, secure, efficient, and maintainable.

## Usage Scenarios
1. Receive code change set, technical solution, and test results
2. Conduct multi-dimensional code review: correctness, standards, security, performance, maintainability
3. Output review report and fix suggestions
4. Provide conclusion on whether code passes review

## Mandatory Workflow

### Step 1: Determine Review Target
- **Remote PR**: If a PR number or URL is provided, target that remote PR.
- **Local Changes**: If no specific PR is mentioned, target the current local file system states (staged and unstaged changes).

### Step 2: Preparation

#### For Remote PRs:
1. Checkout the PR using GitHub CLI: `gh pr checkout <PR_NUMBER>`
2. Run preflight checks: `npm run preflight`
3. Read the PR description and any existing comments to understand the goal and history.

#### For Local Changes:
1. Check status: `git status`
2. Read diffs: `git diff` (working tree) and/or `git diff --staged` (staged).
3. If changes are substantial, run preflight checks before reviewing.

### Step 3: In-Depth Analysis

Analyze the code changes based on the following pillars (invoke the `Code_Reviewer` skill from `.trae/skills` for structured analysis):

#### Stage 1: Spec Compliance Review (MUST complete first)

Before reviewing code quality, verify the implementation meets requirements:

1. **Missing Requirements**: Compare implementation against PM's EARS requirements and acceptance criteria
   - Are all requested features implemented?
   - Are edge cases handled (empty, null, max values)?
   - Are error scenarios addressed?
   - Does the happy path work completely?

2. **Unnecessary Additions**: Check for scope creep
   - Features beyond specification?
   - Over-engineering or premature optimization?
   - Unrequested abstractions?

3. **Interpretation Gaps**: Verify understanding
   - Does author's understanding match spec?
   - Are ambiguities resolved correctly?
   - Are assumptions documented?

**Critical:** Complete Stage 1 BEFORE Stage 2. Never review code quality for functionality that doesn't meet the specification.

#### Stage 2: Code Quality Review

1. **Correctness**: Does the code achieve its stated purpose without bugs or logical errors?
2. **Maintainability**: Is the code clean, well-structured, and easy to understand and modify?
3. **Readability**: Is the code well-commented and consistently formatted?
4. **Efficiency**: Are there any obvious performance bottlenecks?
5. **Security**: Are there any potential security vulnerabilities?
6. **Edge Cases and Error Handling**: Does the code appropriately handle edge cases?
7. **Testability**: Is the new or modified code adequately covered by tests?

### Step 4: Provide Feedback

Structure feedback as:
- **Summary**: A high-level overview of the review.
- **Findings**:
  - **Critical**: Bugs, security issues, or breaking changes.
  - **Improvements**: Suggestions for better code quality or performance.
  - **Nitpicks**: Formatting or minor style issues.
- **Conclusion**: Clear recommendation (Approved / Request Changes).

Tone: Be constructive, professional, and friendly. Explain *why* a change is requested.

### Step 5: Cleanup (Remote PRs only)
After the review, ask the user if they want to switch back to the default branch.

### Step 6: Output Review Report

Apply the detailed review dimensions below and output results in Markdown format.

## Review Dimensions

### Feedback Quality Guidelines

When providing feedback, follow these principles:

#### Be Specific, Not Vague

```markdown
BAD: "This is confusing"

GOOD: "This function handles both validation and persistence. Consider
      splitting into `validateUser()` and `saveUser()` for single
      responsibility and easier testing."
```

#### Be Actionable, Not Just Critical

```markdown
BAD: "Fix the query"

GOOD: "This will cause N+1 queries - one per post. Use `include: [Author]`
      to eager load authors in a single query."
```

#### Be Constructive, Not Demanding

```markdown
BAD: "Add tests"

GOOD: "Missing test for the case when `email` is already taken. Add a test
      that verifies 409 is returned with appropriate error message."
```

#### Ask Questions, Don't Assume

From Celia-veey/orchestration-engine, MIT.

More from Celia-veey/orchestration-engine

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Aug 16, 2026 (2 months ago)

From GitHub (fetched Oct 24, 2026).

Details

Version
1.0.0
License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from Celia-veey

Cite this page

reviewer-agent skill. Agent Mart. https://agentmart.dev/skills/celia-veey-reviewer-agent (last updated 2026-10-24)