refactoring-skills
npx machina-cli add skill wayne930242/Reflexive-Claude-Code/refactoring-skills --openclawRefactoring Skills
Overview
Refactoring skills IS applying clean code principles to documentation.
Consolidate, simplify, extract shared conventions, remove redundancy.
Core principle: Fewer focused skills > many overlapping skills.
Violating the letter of the rules is violating the spirit of the rules.
Task Initialization (MANDATORY)
Before ANY action, create task list using TaskCreate:
TaskCreate for EACH task below:
- Subject: "[refactoring-skills] Task N: <action>"
- ActiveForm: "<doing action>"
Tasks:
- Discover all skills
- Analyze and classify
- Execute refactoring plan
- Extract shared conventions
- Validate results
- Generate report
Announce: "Created 6 tasks. Starting execution..."
Execution rules:
TaskUpdate status="in_progress"BEFORE starting each taskTaskUpdate status="completed"ONLY after verification passes- If task fails → stay in_progress, diagnose, retry
- NEVER skip to next task until current is completed
- At end,
TaskListto confirm all completed
Task 1: Discover All Skills
Goal: Find all skills in the project.
Search locations:
find . -name "SKILL.md" -type f 2>/dev/null
ls -la .claude/skills/ 2>/dev/null
Also check:
- Plugin directories
- User-level skills (
~/.claude/skills/)
Document:
- Skill name
- Location
- Brief description
Verification: Complete list of all skills with locations.
Task 2: Analyze and Classify
Goal: Classify each skill for action.
Classification Matrix
| Status | Criteria | Action |
|---|---|---|
| Keep | Well-structured, unique purpose, follows standards | None |
| Refactor | Has issues but valuable | Use writing-skills to improve |
| Merge | Overlaps with another skill | Consolidate into primary |
| Extract | Has conventions shared by others | Move to rule file |
| Delete | Redundant, unused, or superseded | Remove entirely |
Analysis Questions
For each skill:
- Is the purpose clear and unique?
- Does it overlap with another skill?
- Does it contain shared conventions (→ extract to rule)?
- Is it actually used?
- Does it follow current standards?
Optional: External Search
If claude-skills-mcp is available:
mcp__claude-skills-mcp__search_skills query="[skill domain]"
Compare with community implementations. Better patterns exist?
Verification: Each skill has a classification and action.
Task 3: Execute Refactoring Plan
Goal: Apply the classified actions.
For Skills to Refactor
Invoke the writing-skills skill with the existing skill path.
For Skills to Merge
- Identify the primary skill (keep this one)
- Extract unique content from secondary skill
- Integrate into primary skill
- Delete secondary skill directory
For Skills to Delete
- Confirm no dependencies (grep for skill name)
- Remove entire skill directory
- Remove from any manifest references
CRITICAL: One action at a time. Verify after each.
Verification: All classified actions executed.
Task 4: Extract Shared Conventions
Goal: Move duplicated conventions to rule files.
Signs of shared conventions:
- Same constraint appears in 2+ skills
- Convention is about "what" not "how"
- Convention is path-specific
Process:
- Identify duplicated constraint
- Invoke
writing-rulesskill to create rule - Remove constraint from all skills
- Verify rule auto-injects correctly
Verification: No duplicated conventions across skills.
Task 5: Validate Results
Goal: Verify all remaining skills are valid.
For each skill:
python3 scripts/validate_skill.py <skill-path>
Or manual checklist:
- Frontmatter has name and description
- Description starts with "Use when..."
- Body < 300 lines
- No duplicated conventions (now in rules)
Verification: All remaining skills pass validation.
Task 6: Generate Report
Goal: Document what was done.
Report Format
## Refactoring Summary
### Skills Analyzed: N
### Actions Taken:
| Skill | Action | Notes |
|-------|--------|-------|
| skill-name | kept/refactored/merged/deleted | details |
### Rules Extracted:
| Rule | Source Skills |
|------|--------------|
| rule-name | skill-a, skill-b |
### Recommendations:
- [Future improvements]
Verification: Report accurately reflects all changes.
Red Flags - STOP
These thoughts mean you're rationalizing. STOP and reconsider:
- "Keep all skills, they might be useful"
- "Too much overlap is fine"
- "Don't need to extract conventions"
- "Skip validation, I just fixed them"
- "Report is busywork"
All of these mean: You're about to leave mess behind. Follow the process.
Common Rationalizations
| Excuse | Reality |
|---|---|
| "Keep everything" | Unused skills are clutter. Delete them. |
| "Overlap is fine" | Overlap = confusion. Merge or differentiate. |
| "Skip extraction" | Duplicated conventions drift. Extract to rules. |
| "Skip validation" | Fixed skills can still be broken. Validate. |
| "Skip report" | Report documents decisions for future you. |
Flowchart: Skill Refactoring
digraph refactor_skills {
rankdir=TB;
start [label="Refactor skills", shape=doublecircle];
discover [label="Task 1: Discover\nall skills", shape=box];
analyze [label="Task 2: Analyze\nand classify", shape=box];
execute [label="Task 3: Execute\nrefactoring plan", shape=box];
extract [label="Task 4: Extract\nshared conventions", shape=box];
validate [label="Task 5: Validate\nresults", shape=box];
valid [label="All\nvalid?", shape=diamond];
report [label="Task 6: Generate\nreport", shape=box];
done [label="Refactoring complete", shape=doublecircle];
start -> discover;
discover -> analyze;
analyze -> execute;
execute -> extract;
extract -> validate;
validate -> valid;
valid -> report [label="yes"];
valid -> execute [label="no\nfix"];
report -> done;
}
References
- Use
writing-skillsskill for improving skills - Use
writing-rulesskill for extracting conventions - Use
skill-reviewersubagent for quality checks
Source
git clone https://github.com/wayne930242/Reflexive-Claude-Code/blob/main/plugins/rcc/skills/refactoring-skills/SKILL.mdView on GitHub Overview
Refactoring skills apply clean code principles to project documentation by consolidating and simplifying skills, extracting shared conventions, and removing redundancy. The core principle is that fewer, well-scoped skills outperform many overlapping ones. This skill orchestrates a six-task workflow to discover, classify, refactor, extract conventions, validate, and report results.
How This Skill Works
It orchestrates a six-task workflow using TaskCreate to list actions, then sequentially updates task statuses with TaskUpdate before and after work. It leverages writing-skills for actual refactoring and writing-rules for extracting shared conventions into rule files. Tasks are executed one at a time with verification after each, and a final TaskList confirms completion.
When to Use It
- When skills are messy or overlapping and need consolidation.
- When a project requires a skill review or cleanup triggered by user intent (e.g., 'refactor skills', 'cleanup skills', 'consolidate skills', 'skill review').
- During a formal refactoring or documentation audit to classify each skill (Keep/Refactor/Merge/Extract/Delete).
- Before generating a final skills report or clean manifest to avoid duplicates.
- When external patterns indicate shared conventions that should be moved to rules.
Quick Start
- Step 1: TaskCreate for EACH task below: - Subject: "[refactoring-skills] Task N: <action>" - ActiveForm: "<doing action>"
- Step 2: For each task, run TaskUpdate status="in_progress" before starting, and status="completed" after verification passes
- Step 3: After all tasks, run TaskList to confirm all tasks are completed and generate the final report
Best Practices
- Start by Discovering All Skills to get a complete map.
- Use the Classification Matrix to decide action per skill.
- Refactor one action at a time; verify after each step.
- Extract shared conventions into rule files using writing-rules.
- Validate results with the provided verification steps and run tests (validate_skill).
Example Use Cases
- A project with dozens of overlapping skills is cleaned up to a few focused skills.
- Two nearly identical skills are merged into a primary skill with content extracted.
- Shared conventions like naming schemes moved to a rule file via extraction.
- A task failure triggers diagnostics and retry rather than skipping ahead.
- A final report lists all completed tasks and their locations for auditability.