菜单
Duplicate code detection using similarity-rs. Identifies refactoring candidates based on semantic similarity.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Duplicate code detection using similarity-rs. Identifies refactoring candidates based on semantic similarity.
Run performance benchmarks with criterion. Compare against baselines and profile hotspots.
Implementation verification - runs format, lint, test, and build checks on the workspace.
Production readiness audit - checks unwrap elimination, error handling, clippy, docs, channel usage, and test count.
Release deployment process for ccswarm. Version update, quality gates, build, tag, publish.
Parallel development workflow using git worktrees. Each agent gets an isolated worktree for safe concurrent work.
| name | review-duplicates |
| description | Duplicate code detection using similarity-rs. Identifies refactoring candidates based on semantic similarity. |
| user-invocable | true |
| context | fork |
| agent | code-refactor-agent |
When subagents are enabled in Codex, prefer delegating this workflow to the code-refactor-agent custom agent.
Detect duplicate code in the ccswarm codebase using similarity-rs.
similarity-rs crates/ --threshold 0.80
For Rust code, also consult cargo coupling so duplicate cleanup is prioritized
by architectural impact, not only by textual similarity.
cargo coupling . --summary --exclude-tests
cargo coupling . --hotspots=12 --exclude-tests
If a duplicate cluster is located in a coupling hotspot, high-risk boundary, or cycle-related module, raise its refactoring priority accordingly.
Classify results by priority:
| Priority | Condition | Action |
|---|---|---|
| High | 95%+ similarity, 10+ lines | Extract to common function |
| Medium | 90-95% similarity, 5+ lines | Consider trait or macro |
| Low | 85-90% similarity | Evaluate structural similarity |
| Skip | Test code, config structs | Allow intentional duplication |
Output a JSON report with high_priority, medium_priority, low_priority, and ignored sections.