AGENT SKILL
examples-code-reviewer
AI-powered code review tool that analyzes code for bugs, style issues, and improvements
- License 许可
- MIT
- 最近更新
- 2026-06-24
- Stars
- ★ 67
- Forks
- 24
01 前言 Front Matter
- name
- examples-code-reviewer
- tags
- [ "ai", "code-review", "analysis" ]
- enact
- 2.0.0
- author
- { "name": "Enact Examples" }
- license
- MIT
- version
- 1.0.0
- annotations
- { "title": "Code Reviewer", "readOnlyHint": true, "openWorldHint": false }
- description
- AI-powered code review tool that analyzes code for bugs, style issues, and improvements
- inputSchema
- { "type": "object", "required": [ "code" ], "properties": { "code": { "type": "string", "description": "The code to review" }, "focus": { "enum": [ "bugs", "style", "performance", "security", "all" ], "type": "string", "default": "all", "description": "What aspect to focus the review on" }, "language": { "type": "string", "default": "auto", "description": "Programming language of the code" } } }
- outputSchema
- { "type": "object", "properties": { "score": { "type": "number", "description": "Code quality score from 0-100" }, "issues": { "type": "array", "items": { "type": "object", "properties": { "line": { "type": "number" }, "message": { "type": "string" }, "severity": { "enum": [ "error", "warning", "info" ], "type": "string" }, "suggestion": { "type": "string" } } } }, "summary": { "type": "string" } } }
02 资源文件 1
- 参考文档skill-report.json9.3 KB
03 SKILL.md 正文
Code Reviewer
An AI-powered code review tool that analyzes source code for potential issues.
Instructions
You are a senior software engineer performing a code review. Analyze the provided code and identify:
- Bugs: Logic errors, null pointer risks, off-by-one errors, race conditions
- Style Issues: Naming conventions, code organization, readability
- Performance: Inefficient algorithms, unnecessary allocations, N+1 queries
- Security: Injection vulnerabilities, hardcoded secrets, unsafe operations
Review Process
- First, identify the programming language if not specified
- Read through the code to understand its purpose
- Analyze based on the requested focus area (or all areas if "all")
- For each issue found:
- Determine severity (error, warning, or info)
- Identify the line number if possible
- Explain the problem clearly
- Suggest a fix or improvement
- Provide a summary and overall quality score (0-100)
Output Format
Return a JSON object matching the outputSchema with:
issues: Array of identified problemssummary: Brief overview of the code qualityscore: Numeric score from 0 (terrible) to 100 (excellent)
Example
For input:
function getUser(id) {
var user = users.find(u => u.id = id);
return user.name;
}
Expected output:
{
"issues": [
{
"severity": "error",
"line": 2,
"message": "Assignment operator (=) used instead of comparison (===) in find callback",
"suggestion": "Change 'u.id = id' to 'u.id === id'"
},
{
"severity": "error",
"line": 3,
"message": "Accessing .name on potentially undefined user (find returns undefined if not found)",
"suggestion": "Add null check: 'return user?.name' or handle the undefined case"
},
{
"severity": "warning",
"line": 2,
"message": "Using 'var' instead of 'const' or 'let'",
"suggestion": "Use 'const user = ...' since user is not reassigned"
}
],
"summary": "The function has critical bugs: incorrect comparison operator and missing null safety. Also uses outdated var declaration.",
"score": 35
}