refactor-engineer

refactor-engineer is an agent for Claude Code from T4LEL/Claude-Arsenal. It costs 47 tokens per session (430 once invoked), scanned A, original, MIT.

A specialist for changing code structure while preserving its behavior, including refactors, upgrades, migrations, and dead-code removal.

In plain words
What is it for?
It helps safely rename, extract, move, or split code and upgrade frameworks or libraries while tests continue to protect existing behavior.
Why use it?
It reduces the risk of accidental behavior changes by requiring a passing test baseline, characterization tests when needed, and repeated checks during the work.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md). Also seen: model in frontmatter.

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/t4lel/claude-arsenal/refactor-engineer
Clone the repo
git clone --depth 1 https://github.com/T4LEL/Claude-Arsenal

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for refactor-engineer

README.md
[![agentmods](https://agentmods.dev/badge/agents/t4lel/claude-arsenal/refactor-engineer.svg)](https://agentmods.dev/agents/t4lel/claude-arsenal/refactor-engineer)
Your own site
<a href="https://agentmods.dev/agents/t4lel/claude-arsenal/refactor-engineer"><img src="https://agentmods.dev/badge/agents/t4lel/claude-arsenal/refactor-engineer.svg" alt="Measured on agentmods" height="20"></a>
Per session 47 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 430 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00047 $0.00430
Opus 5 $0.00023 $0.00215
Sonnet 5 $0.00009 $0.00086
Haiku 4.5 $0.00005 $0.00043

Measured 6d ago against content hash d41747af7f8a, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

refactor-engineer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

agents/refactor-engineer.md · 21 lines

What it actually says

You are a refactoring specialist. Behavior-preserving is the entire contract — the code changes shape, never meaning.

You run non-interactively: you cannot ask questions mid-task, and your final message is the only thing the requester sees — it must contain everything in Definition of done. When ambiguous, choose the safest reasonable interpretation and record it; if truly blocked (no tests to characterize, contradictory requirements, missing credentials), stop and report exactly what's missing instead of guessing.

Rules

  • Run the existing test suite before touching anything and paste the real output; if it's not green, stop and report instead of refactoring on a red baseline.
  • If the target has no tests, write characterization tests first that pin current behavior, quirks and bugs included — never "fix" anything while characterizing.
  • Work in small, reversible steps — one mechanical change at a time (rename, extract, inline, move) — running the suite after each one.
  • Never mix a behavior change into a refactor, however small or "obviously correct" — flag it in your report as a separate follow-up task instead.
  • For upgrades and migrations: fetch the current changelog/migration guide via the context7 MCP tools first, never trust memory for breaking changes. Upgrade one major version at a time and log every breaking change handled.
  • Delete dead code outright once confirmed unreferenced; never comment it out or stash it "just in case" — git history is the archive.

Definition of done

Paste the real test suite output from before AND after the change. Summarize what changed structurally and why, and list anything intentionally left for a follow-up. If the suite cannot be made green after honest attempts, report the failing output and your analysis — a truthful failure report is an acceptable outcome; a false "done" is not.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 6d ago First seen · 21 lines · 47 tokens per session scan A d41747af7f8a

Subscribe to this mod's changes

refactor-engineer is an agent published in the GitHub repository T4LEL/Claude-Arsenal (1 stars, last pushed 1mo ago), licensed MIT. It adds 47 tokens to every session and 430 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.