compliance-review
This skill should be used when the user asks to "check CLAUDE.md compliance", "review against coding standards", "check AI agent instructions", "verify guidelines", "check coding conventions", "check naming conventions", or wants to ensure code follows project-specific rules and standards.
安装与使用
请根据来源仓库和你的 AI 工具说明安装此技能。先阅读技能文件和依赖要求,再添加到工作流。
查看来源 ↗使用指南技能说明
Compliance Code Review Skill
Verify code adherence to AI Agent Instructions (CLAUDE.md, copilot-instructions, and similar files) and project-specific rules through targeted compliance-focused code review.
Agent Configuration
Uses compliance-agent (Sonnet in all modes). See ${CLAUDE_PLUGIN_ROOT}/shared/review-workflow.md for authoritative model configuration.
Instruction File Locations (Part of Step 2)
See ${CLAUDE_PLUGIN_ROOT}/shared/skill-common-workflow.md Step 2 for the canonical search order.
Rule Classification
For rule classification (MUST/SHOULD/MAY keywords and severity mapping), see references/compliance-patterns.md.
Compliance Categories Checked
Architecture Violations (Major):
- Layer violations (UI importing from data layer)
- Business logic in wrong location (controller, view)
- Direct database access bypassing repositories
- Circular dependencies
Naming Convention Violations (Minor to Major):
- File naming patterns (kebab-case, PascalCase)
- Class/function/variable naming
- Constant naming (UPPER_SNAKE_CASE)
- Interface prefixing (I prefix in C#)
Documentation Requirements (Minor):
- Missing JSDoc/XMLDoc on public APIs
- Undocumented complex functions
- README updates for new features
Security Requirements (Major):
- Missing authentication on endpoints
- Input validation gaps
- Logging requirements for sensitive operations
Auto-Validated Patterns
High-confidence patterns that skip validation. For full definitions, see ${CLAUDE_PLUGIN_ROOT}/shared/validation-rules.md.
Compliance patterns: missing_authorize_attribute, wrong_case_filename, explicit_must_violation, missing_required_jsdoc
False Positives
Apply all rules from ${CLAUDE_PLUGIN_ROOT}/shared/false-positives.md.
Compliance-specific additions - do NOT flag:
- Explicit override comments (
// claude-ignore: rule-name) - Scope mismatch (rule doesn't apply to this file type)
- Ambiguous rules (benefit of doubt to code)
- Framework exemptions (framework provides required behavior)
Rule Citation Guidelines
Always include exact rule citation:
- Rule text: Exact quoted text from instruction file
- Source: File path and line number
- Keyword: The strength keyword (MUST, SHOULD, MAY)
Example: "The endpoint violates 'All API endpoints MUST have authentication' from CLAUDE.md:31"
Compliance Score
Include in output:
### Compliance Score
- Rules checked: 28
- Rules passed: 24
- Rules violated: 4
- Compliance rate: 86%
Example Output
See examples/example-output.md for a sample showing:
- Missing authentication violation with diff fix
- Business logic in controller with prompt fix
- Missing JSDoc violation with diff fix
Related Components
See ${CLAUDE_PLUGIN_ROOT}/agents/compliance-agent.md for the agent definition.
使用示例
查看来源仓库中的使用示例与配置说明。
查看来源 ↗来源历史
- 收录时间
- 2026-01-20
- 索引更新时间
- 2026-01-26
- Branch
- main
收录与同步日期属于本站记录。源码修改历史请查看来源仓库。
GitHub ↗chat Comments (0)
Sign in to join the discussion and leave a comment.
No comments yet. Be the first to share your thoughts!