Refactor Executor — Testing & QA agent for Claude Code
Use when a refactor spec from spec-writer is ready to execute.
How to install Refactor Executor
Installs to ~/.claude/agents/onlygian-g-forge-refactor-executor.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/onlygian/G-Forge/HEAD/agents/refactor-executor.md -o ~/.claude/agents/onlygian-g-forge-refactor-executor.md Restart Claude Code, or start a new session, for it to be picked up.
What Refactor Executor does
name: refactor-executor description: Use when a refactor spec from spec-writer is ready to execute. Executes exactly as written — no scope creep, no judgment calls. model: haiku tools: Read, Glob, Grep, Write, Edit, Bash color: green maxTurns: 20 isolation: worktree
You execute refactor specs exactly as written. You do not interpret, improve, or expand scope.
Input
A refactor spec from spec-writer containing: goal, files to touch, explicit steps, done condition.
Execution rules
Alternatives in Testing & QA
- Unit Executor — Implements exactly one unit of work from an inline spec and returns a structured JSON record 314 ★
- Code Reviewer Spec — Use this agent to verify that a pull request or set of changes fully implements its stated requirements — chec 186 ★
- App Operator — 在已有 Marionette 兼容 MCP 时,只对通过静态审计的 ready UI 行为 Spec 执行运行态验证并写报告;不生成 Spec、不准备环境、不探索流程或修改代码 113 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Wasp Refactor Executor
Execute mechanical refactorings (DRY, extract function, rename) while maintaining test green status. Optimized
Vela Executor
Vela 실행자 — TDD 방식(test-write→implement→refactor)으로 plan.md를 구현한다. PM이 execute 단계에서 Agent 도구로 호출한다.
Fe Analyst
Requirement gap analysis before/during the fe-spec stage. Systematically identifies unasked questions, undefin
Build Reviewer
Reviews built work on two axes (code standards and spec conformance), runs the scope-creep check against the p
Sms Assistant Test Writer
Use this agent when the user needs to write or improve unit tests for the SMS assistant functionality, particu
Builder Slim
Execute plan steps exactly with dry-run and test support
Related Skills
Agent Context Framework
Production-ready Markdown (.md) context templates for AI coding agents (Cursor, Claude Code, Aider, Windsurf).
Ext Exploit
Execute ONE approved step from the external attack plan via the ext-exploit-operator agent, with human confirm
Session Planner
You are the Session Planner Wizard. Your job is to interview the user and generate a complete, ready-to-use SE