Install
$ agentstack add skill-pradnyeshp-claude-skills-refactor ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Test setup (used to verify behavior is preserved)
!ls package.json pyproject.toml Cargo.toml go.mod pom.xml build.gradle Makefile 2>/dev/null
Instructions
Refactor the target named in $ARGUMENTS toward the stated goal. If the target or goal is unclear, ask which code to refactor and what outcome they want before touching anything.
Core principle
Behavior must not change. A refactor changes how code is expressed, never what it does. Public APIs, return values, and side effects stay identical unless the user explicitly asks otherwise.
Method
- Read first. Open the target and its call sites (use Grep). Understand the current behavior and who depends on it before changing it.
- Establish a safety net. Identify the tests covering this code. If none exist and the change is non-trivial, say so and offer to add characterization tests first (the
test-genskill can help). - Make small, reviewable moves toward the goal: extract functions/variables, rename for clarity, remove duplication, simplify conditionals, separate concerns. Keep each change focused.
- Update all call sites and imports affected by the change. Don't leave the codebase in a broken intermediate state.
- Verify. Run the relevant tests / type-checker / linter and confirm they still pass.
Rules
- Preserve behavior, signatures, and side effects. Flag explicitly if a change is unavoidably behavior-altering.
- Don't gold-plate — make the change the goal requires, not a sweeping rewrite of unrelated code.
- Match the project's existing style and idioms; don't introduce new patterns or dependencies just to refactor.
- Work in coherent steps and explain each. If the refactor is large, outline the plan and confirm before applying.
- End with a summary of what moved/changed and the verification result.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: Pradnyeshp
- Source: Pradnyeshp/Claude-Skills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.