refactoring
Refactor Code in Small Verified Steps
Large refactors can change behavior and hide regressions. This skill guides Claude, Codex, and Claude Code through small tested changes with commits after each safe step.
Install with my Agent
Copy this request to your Agent. It includes the canonical Skill page and manifest.
Review the Skillstore skill "refactoring" from https://skillstore.io/skills/arcadeai-refactoring.md and its manifest at https://skillstore.io/api/skills/arcadeai-refactoring/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.Your Agent should still show its plan and request any confirmation required by the security policy.
Agent-readable resources
Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.
Test it
Using "refactoring". Refactor this discount function.
Expected outcome:
- Classifies the request as refactoring, not feature work.
- Checks or adds tests for current discount behavior.
- Renames unclear values and runs tests before committing.
Using "refactoring". Clean up this class with too many responsibilities.
Expected outcome:
- Finds the smallest extractable responsibility.
- Moves one method or group at a time.
- Stops after each passing test run for review.
Using "refactoring". Tests failed after the refactor.
Expected outcome:
- Stops the refactoring workflow.
- Explains that behavior may have changed.
- Asks how to proceed before further edits.
Security Audit
Medium RiskMost static findings are false positives caused by Markdown fences and inline code examples, not executable backtick shell usage. No evidence of prompt injection, network access, credential access, or data exfiltration was found. Two semantic risks remain: wildcard tool permissions and revert guidance that could discard uncommitted work without confirmation.
Confirmed security concerns (2)
Risk Factors
⚙️ External commands (18)
Share & cite this report
Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.
Copy report link
https://skillstore.io/skills/arcadeai-refactoring/audits/8?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown badge
[](https://skillstore.io/skills/arcadeai-refactoring?utm_source=security_passport_badge)HTML badge
<a href="https://skillstore.io/skills/arcadeai-refactoring?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/arcadeai-refactoring/security.svg" alt="Skillstore security assessment" loading="lazy"></a>Embed card
<iframe src="https://skillstore.io/embed/skills/arcadeai-refactoring.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>Academic citations (APA · BibTeX · CFF)
APA citation
ArcadeAI. (2026). refactoring security audit report (audit version 8) [Author version unspecified]. Skillstore. https://skillstore.io/skills/arcadeai-refactoring/audits/8BibTeX citation
@techreport{arcadeai-arcadeai-refactoring-2026,
author = {ArcadeAI},
title = {refactoring security audit report (audit version 8)},
institution = {Skillstore},
year = {2026},
number = {8},
url = {https://skillstore.io/skills/arcadeai-refactoring/audits/8},
note = {Author version unspecified}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "refactoring security audit report (audit version 8)"
version: "unspecified"
type: report
authors:
- name: "ArcadeAI"
date-released: "2026-07-05"
url: "https://skillstore.io/skills/arcadeai-refactoring/audits/8"
identifiers:
- type: other
value: "skillstore:arcadeai-refactoring:audit:8"
description: "Skillstore immutable audit report identifier"
Skillstore Score
Why this score Evidence Confidence: HighWhat You Can Build
Clean Up Legacy Functions
Improve unclear names, long functions, and nested conditionals while preserving current behavior.
Prepare Code Before Feature Work
Reduce technical debt before adding new behavior, using tests to protect existing outcomes.
Teach Safe Refactoring Discipline
Guide engineers through small changes, test checks, and clear commits after each successful step.
Try These Prompts
Use the refactoring skill on this function. First assess whether it is refactoring, then make one safe change and run tests.
This module has limited tests. Add characterization tests for the behavior you will touch, then complete the smallest useful refactor.
Refactor this class without changing behavior. Break the work into separate rename, extract, or move steps, testing after each step.
Analyze this large refactoring request. Propose the smallest safe sequence, identify test gaps, and stop before any behavior-changing work.
Best Practices
- Keep each refactoring small enough to explain in one sentence.
- Add characterization tests before changing code with weak coverage.
- Review commits after each passing test run.
Avoid
- Combining renames, extracts, and behavior changes in one edit.
- Treating failed tests as bugs to fix inside the refactor.
- Running destructive revert commands without confirming affected files.