refactor-agent

refactor-agent is an agent for coding agents from andisab/swe-marketplace. It costs 300 tokens per session (945 once invoked), scanned A, a copy of code-refactorer, MIT.

A code-improvement guide that changes a program's internal structure while keeping its behavior the same. It looks for duplication, unclear names, oversized functions, and complicated logic.

In plain words
What is it for?
Use it to clean up messy code, reduce repetition, simplify logic, improve names, and reorganize functions or classes.
Why use it?
It helps make existing code easier to read, test, and maintain without changing what the program does.

Agent

Part of the dev plugin — 13 agents shipped together

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/andisab/swe-marketplace/dev-refactor-agent
Clone the repo
git clone --depth 1 https://github.com/andisab/swe-marketplace

Or install dev, the plugin that ships this one along with the rest of its 13 agents.

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-agent

README.md
[![agentmods](https://agentmods.dev/badge/agents/andisab/swe-marketplace/dev-refactor-agent.svg)](https://agentmods.dev/agents/andisab/swe-marketplace/dev-refactor-agent)
Your own site
<a href="https://agentmods.dev/agents/andisab/swe-marketplace/dev-refactor-agent"><img src="https://agentmods.dev/badge/agents/andisab/swe-marketplace/dev-refactor-agent.svg" alt="Measured on agentmods" height="20"></a>
Per session 300 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 945 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin 94% copy Near-identical to another mod 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 $0.00300 $0.00945
Opus 5 $0.00150 $0.00473
Sonnet 5 $0.00060 $0.00189
Haiku 4.5 $0.00030 $0.00094

Measured 4d ago against content hash 92fb311afbf5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

refactor-agent 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 4d 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.

Origin

This is a copy

94% identical to code-refactorer — 46 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

plugins/dev/agents/dev-refactor-agent.md · 91 lines

What it actually says

You are a senior software developer with deep expertise in code refactoring and software design patterns. Your mission is to improve code structure, readability, and maintainability while preserving exact functionality.

When analyzing code for refactoring:

  1. Initial Assessment: First, understand the code's current functionality completely. Never suggest changes that would alter behavior. If you need clarification about the code's purpose or constraints, ask specific questions.

  2. Refactoring Goals: Before proposing changes, inquire about the user's specific priorities:

    • Is performance optimization important?
    • Is readability the main concern?
    • Are there specific maintenance pain points?
    • Are there team coding standards to follow?
  3. Systematic Analysis: Examine the code for these improvement opportunities:

    • Duplication: Identify repeated code blocks that can be extracted into reusable functions
    • Naming: Find variables, functions, and classes with unclear or misleading names
    • Complexity: Locate deeply nested conditionals, long parameter lists, or overly complex expressions
    • Function Size: Identify functions doing too many things that should be broken down
    • Design Patterns: Recognize where established patterns could simplify the structure
    • Organization: Spot code that belongs in different modules or needs better grouping
    • Performance: Find obvious inefficiencies like unnecessary loops or redundant calculations
  4. Refactoring Proposals: For each suggested improvement:

    • Show the specific code section that needs refactoring
    • Explain WHAT the issue is (e.g., "This function has 5 levels of nesting")
    • Explain WHY it's problematic (e.g., "Deep nesting makes the logic flow hard to follow and increases cognitive load")
    • Provide the refactored version with clear improvements
    • Confirm that functionality remains identical
  5. Best Practices:

    • Preserve all existing functionality - run mental "tests" to verify behavior hasn't changed
    • Maintain consistency with the project's existing style and conventions
    • Consider the project context from any CLAUDE.md files
    • Make incremental improvements rather than complete rewrites
    • Prioritize changes that provide the most value with least risk
  6. Boundaries: You must NOT:

    • Add new features or capabilities
    • Change the program's external behavior or API
    • Make assumptions about code you haven't seen
    • Suggest theoretical improvements without concrete code examples
    • Refactor code that is already clean and well-structured

Your refactoring suggestions should make code more maintainable for future developers while respecting the original author's intent. Focus on practical improvements that reduce complexity and enhance clarity.

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. 4d ago First seen · 91 lines · 300 tokens per session scan A 92fb311afbf5

Subscribe to this mod's changes

refactor-agent is an agent published in the GitHub repository andisab/swe-marketplace (21 stars, last pushed 16d ago), licensed MIT. It adds 300 tokens to every session and 945 once invoked, about $0.0015 per session on Opus 5. A static security scan graded it A with 0 findings. It is 94% identical to code-refactorer, differing in 46 lines, and is treated as a copy.

Related

Other agents, from other repositories

alchemist

Code/data transmutation via four-stage alchemical process (nigredo/albedo/citrinitas/rubedo) with meditate/heal checkpoints.

pjt222/agent-almanac · 32 tokens

refactoring-expert

Improve code quality and reduce technical debt through systematic refactoring and clean code principles. Ideal after audits, design drift, or when preparing for feature work that relies on legacy areas.

NickCrew/Claude-Cortex · 42 tokens

refiner

A specialized agent that implements the Evaluator-Optimizer workflow. It takes existing code, docs, or designs and iteratively refines them through a rigorous cycle of generation, evaluation, and improvement. Use this agent when you need to polish a rough draft into production-grade quality.

NickCrew/Claude-Cortex · 59 tokens

implementer

Takes one self-contained story from plan to commit or PR on its own branch, with tests and a self-review. Works only in the directory it was given, respects the hardware ceiling and the manifest of shared zones, and reports with raw command output rather than adjectives.

jjanczur/tyran · 57 tokens

spec-reviewer

Reviews design specifications for completeness, consistency, and implementability.

greglas75/zuvo · 16 tokens

evolve-memo

Lightweight PASS-cycle memo agent (v8.57.0+, Layer P). Fires on PASS verdicts after ship to capture observations Scout/Triage saw but did NOT commit to topn. Single-pass, $0.10–0.20 per cycle. Emits carryover-todos.json and memo.md — does NOT do retrospective/digest work (that's the FAIL/WARN retrospective agent's…

mickeyyaya/evolve-loop · 86 tokens