From code
Team-based code refactoring with analysis and user-approved changes. Use when: "refactor this code", "リファクタリング", "コード整理", "refactor team", "リファクタチーム".
npx claudepluginhub signalcompose/claude-tools --plugin codeThis skill uses the workspace's default tool permissions.
Resolve target files:
Generates design tokens/docs from CSS/Tailwind/styled-components codebases, audits visual consistency across 10 dimensions, detects AI slop in UI.
Records polished WebM UI demo videos of web apps using Playwright with cursor overlay, natural pacing, and three-phase scripting. Activates for demo, walkthrough, screen recording, or tutorial requests.
Delivers idiomatic Kotlin patterns for null safety, immutability, sealed classes, coroutines, Flows, extensions, DSL builders, and Gradle DSL. Use when writing, reviewing, refactoring, or designing Kotlin code.
Resolve target files:
git diff --name-only <base-branch> for changed filesDetect context:
git symbolic-ref refs/remotes/origin/HEAD 2>/dev/null | sed 's|refs/remotes/origin/||' (fallback: main)package.json, Makefile, pytest.ini, etc.MANDATORY: Use TeamCreate to create the team, then spawn agents with Task tool.
Team structure:
Team Lead (yourself)
├─ Analyzer: pr-review-toolkit:code-simplifier
└─ Refactorer: general-purpose (refactorer)
MANDATORY: Always specify an explicit model parameter when spawning each agent. Choose the appropriate model based on task complexity (haiku for lightweight, sonnet for standard, opus for complex reasoning). Never omit model (default inherit may fail in parallel spawning).
Analyzer examines target files using criteria from ${CLAUDE_PLUGIN_ROOT}/skills/refactor-team/references/analysis-criteria.md.
Analysis output format:
| # | Category | File:Line | Description | Priority | Risk |
|---|---|---|---|---|---|
| 1 | DRY | src/a.ts:42 | ... | High | Low |
| 2 | Naming | src/c.ts:8 | ... | Medium | None |
MANDATORY: Present the analysis table to user and wait for approval. Do NOT proceed without explicit user selection of which items to execute.
User selects items by number (e.g., "1, 3, 5" or "all").
For each approved item:
refactor: <description>Rules:
refactor:Report summary:
Send shutdown_request to all agents individually via SendMessage (one message per agent):
Wait up to 30 seconds for shutdown_response from each agent.
Then call TeamDelete.
If TeamDelete fails (agents did not respond to shutdown):
dangerouslyDisableSandbox: true:
rm -rf ~/.claude/teams/<team-name>/ ~/.claude/tasks/<team-name>/