القائمة
Plan a multi-file Rust refactor safely: investigate, plan, confirm, implement. Use before any multi-file refactor.
يُحدَّد Prompt الذي يراجع المصدر أولًا بشكل افتراضي. يمكنك التبديل إلى أمر مباشر أو تنزيل نسخة محلية.
اقرأ SKILL.md وأي ملفات مرافقة يعرضها SkillsMP قبل أن تقرر التثبيت.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
يتجاوز الأمر المباشر Prompt المخصّص للمراجعة. افحص المصدر قبل تشغيله.
npx skills add https://github.com/Cosmian/kms --skill refactor-planيبقى الأمر في سطر واحد. مرّر أفقيًا لمراجعته كاملًا قبل النسخ.
تفضّل نسخة محلية؟ نزّل الملفات المتاحة حاليًا لدى SkillsMP.
Scan Rust code for every form of panic or brutal process exit: panic!, todo!, unimplemented!, unreachable!, .unwrap(), .expect() in production, process::exit/abort, integer overflow, and out-of-bounds indexing. Produces a ranked report with patches. Use when reviewing Rust code quality, before any PR, or when adding new Rust code.
Hardcore Rust code review gate: orchestrates ALL Rust review skills in sequence (panic audit, error propagation, async refactor, simplify, refactor, patterns, security, cryptography). Each skill writes its report to ./review/. Produces a unified go/no-go verdict. Use before submitting any significant Rust PR or after large code generation.
Write the next version entry at the top of CHANGELOG.md by summarizing all changes since the last tagged release. Use when preparing release notes.
استنادا إلى تصنيف SOC المهني
عرض SKILL.md
| name | refactor-plan |
| description | Plan a multi-file Rust refactor safely: investigate, plan, confirm, implement. Use before any multi-file refactor. |
Create a detailed, safe plan before making any code changes in this Rust codebase.
cargo commands.If the request is too ambiguous to plan safely, ask concise clarifying questions instead of editing files.
Use these at each phase checkpoint:
cargo clippy-all # zero warnings required
cargo fmt --all # formatting
cargo test -p <crate> # targeted test (preferred)
cargo test-fips # full FIPS test suite (only if needed)
cargo test-non-fips # full non-FIPS suite (only if needed)
Use cargo test -p <crate> <test_name> to run the narrowest scope first.
## Refactor Plan: [title]
### Current State
[How things work now — be specific to this codebase]
### Target State
[How things will work after — measurable outcome]
### Affected Files
| File | Change Type | Dependencies |
|------|-------------|--------------|
| `crate/foo/src/bar.rs` | modify | blocks `baz.rs`, blocked by `qux.rs` |
### Execution Plan
#### Phase 1: Types and Traits
- [ ] Step 1.1: [action] in `crate/foo/src/bar.rs`
- [ ] Verify: `cargo clippy -p foo` passes
#### Phase 2: Implementations
- [ ] Step 2.1: [action] in `crate/foo/src/impl.rs`
- [ ] Verify: `cargo test -p foo test_name`
#### Phase 3: Callers
- [ ] Step 3.1: Update callers in `crate/server/src/...`
- [ ] Verify: `cargo build --workspace`
#### Phase 4: Tests
- [ ] Step 4.1: Update unit tests
- [ ] Verify: `cargo test -p foo`
#### Phase 5: Cleanup
- [ ] Remove deprecated code
- [ ] Run `cargo fmt --all` and `cargo clippy-all`
- [ ] Final verify: `git diff --stat` — every hunk explainable by the task
### Rollback Plan
If Phase N fails:
1. `git checkout crate/foo/src/bar.rs` — restore original
2. `git stash` — save partial work for inspection
### Risks
- [Potential issue]: [mitigation]
- Trait object overhead: prefer generics if the type set is closed
Public API change: run across workspace before deleting
After the plan, ask: "Shall I proceed with Phase 1?"