Skill

refactor-classification-policy

Classifies RTL refactoring changes into SAFE/RESTRICTED/PROHIBITED tiers with approval requirements for review-refactor workflows.

code-quality

Install

npx claudepluginhub babyworm/rtl-agent-team --plugin rtl-agent-team

Tool Access

This skill uses the workspace's default tool permissions.

Preview

- `SAFE`: behavior-preserving cleanup with no interface/timing contract impact

SKILL.md

Similar Skills

design-system

Generates design tokens/docs from CSS/Tailwind/styled-components codebases, audits visual consistency across 10 dimensions, detects AI slop in UI.

team-skills-platform

163.7k

ui-demo

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.

team-skills-platform

163.7k

kotlin-patterns

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.

team-skills-platform

163.7k

Stats

Stars12

Forks3

Last CommitMar 19, 2026

Actions

View Source View Plugin View on GitHub View README

Refactor Policy

Refactor Classes

SAFE: behavior-preserving cleanup with no interface/timing contract impact
RESTRICTED: likely behavior-preserving but affects structure or test assumptions
PROHIBITED: behavior/spec contract may change

SAFE (auto-apply allowed)

Naming/format consistency fixes
Dead code removal
Local duplication cleanup with equivalent logic
Comment/documentation synchronization

RESTRICTED (approval or explicit orchestrator permit)

Significant module decomposition/merge
Testbench structure reshaping affecting check points
Constraint file reorganization without semantic change proof

PROHIBITED (manual approval required)

FSM semantic changes
Pipeline latency/stage changes
CDC strategy changes
Interface contract changes
Reset behavior semantics changes

Pass/Fail Gate

FAIL: any prohibited change applied without approval
FAIL: equivalence confidence not demonstrated for SAFE/RESTRICTED changes
PASS: only SAFE changes applied and mandatory recheck passed

Escalation

Any ambiguity about behavior equivalence: escalate as RESTRICTED
Any proposal touching interface/timing contract: escalate before edit
Repeated recheck fail after 2 attempts: stop and hand off with root-cause note

Output Format

Use the following refactor execution report:

# Refactor Execution Report
- Scope: module|block|top
- Verdict: PASS | FAIL

## Planned Changes
| ID | Class | Target | Rationale | Approval Required |
|---|---|---|---|---|

## Applied Changes
| ID | Files | Summary | Risk |
|---|---|---|---|

## Blocked/Deferred
- [approval-required or risky items]