Code Refiner
Mathews-Tom/armory
Deep code simplification and refactoring preserving behavior across Python, Go, TypeScript, Rust.
Find and safely apply per-operator refactoring / redundancy-reduction opportunities in a CV-CUDA operator (near-duplicate Tensor/VarShape kernels, reinvented shared utilities, dead code).
Install Refactor Op
$ npx skills add CVCUDA/CV-CUDA --skill refactor-op -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install CVCUDA/CV-CUDA refactor-op --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/CVCUDA/CV-CUDA.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/refactor-op .claude/skills/refactor-op && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "refactor-op" agent skill from https://github.com/CVCUDA/CV-CUDA/tree/main/.agents/skills/refactor-op into .claude/skills/refactor-op/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "refactor-op", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/CVCUDA/CV-CUDA/tree/main/.agents/skills/refactor-opType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add CVCUDA/CV-CUDA --skill refactor-op -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install CVCUDA/CV-CUDA refactor-op --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/CVCUDA/CV-CUDA.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.agents/skills/refactor-op .agents/skills/refactor-op && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "refactor-op" agent skill from https://github.com/CVCUDA/CV-CUDA/tree/main/.agents/skills/refactor-op into .agents/skills/refactor-op/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "refactor-op", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add CVCUDA/CV-CUDA --skill refactor-op -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install CVCUDA/CV-CUDA refactor-op --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/CVCUDA/CV-CUDA.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.agents/skills/refactor-op .cursor/skills/refactor-op && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "refactor-op" agent skill from https://github.com/CVCUDA/CV-CUDA/tree/main/.agents/skills/refactor-op into .cursor/skills/refactor-op/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "refactor-op", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/CVCUDA/CV-CUDA.git --path .agents/skills/refactor-op--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add CVCUDA/CV-CUDA --skill refactor-op -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install CVCUDA/CV-CUDA refactor-op --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/CVCUDA/CV-CUDA.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.agents/skills/refactor-op .gemini/skills/refactor-op && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "refactor-op" agent skill from https://github.com/CVCUDA/CV-CUDA/tree/main/.agents/skills/refactor-op into .gemini/skills/refactor-op/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "refactor-op", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install CVCUDA/CV-CUDA refactor-opInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add CVCUDA/CV-CUDA --skill refactor-op -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/CVCUDA/CV-CUDA.git skills-src && mkdir -p .github/skills && cp -r skills-src/.agents/skills/refactor-op .github/skills/refactor-op && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "refactor-op" agent skill from https://github.com/CVCUDA/CV-CUDA/tree/main/.agents/skills/refactor-op into .github/skills/refactor-op/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "refactor-op", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add CVCUDA/CV-CUDA --skill refactor-op -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install CVCUDA/CV-CUDA refactor-op --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/CVCUDA/CV-CUDA.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.agents/skills/refactor-op .opencode/skills/refactor-op && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "refactor-op" agent skill from https://github.com/CVCUDA/CV-CUDA/tree/main/.agents/skills/refactor-op into .opencode/skills/refactor-op/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "refactor-op", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
refactor-op- GitHub stars
- 2.7k
- Token cost
- ~1.5k tokens
- SKILL.md length
- 640 words
- Files
- 1
- Skills in repo
- 12
- Repo updated
- First seen
- Licence
- Custom licence
At a glance
Find and safely apply per-operator refactoring / redundancy-reduction opportunities in a CV-CUDA operator (near-duplicate Tensor/VarShape kernels, reinvented shared utilities, dead code).
- Works in 3 steps: assess — python3 tools/refactor_op.py… → apply (only if asked to fix) — apply the… → verify — when the MR scope includes a…
- Asked to reduce code duplication
- SKILL.md covers Workflow and Prompt Handling
- Calls python3
What it does
Refactor Op is an agent skill from CVCUDA/CV-CUDA. Find and safely apply per-operator refactoring / redundancy-reduction opportunities in a CV-CUDA operator (near-duplicate Tensor/VarShape kernels, reinvented shared utilities, dead code). Use when asked to reduce code duplication, de-duplicate or unify an operator's kernels/bindings, remove dead code, or verify a refactor changed nothing observable. Produces a deterministic findings-first report; the apply path is gated by a strict bit-exact / feature-set / test-coverage parity check.
Its SKILL.md is about 1.5k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It sits in Development, covering Refactoring, Code simplification and Test coverage. It works with CUDA, NVIDIA AI Platform and Python. The repository describes itself as: CV-CUDA™ is an open-source, GPU accelerated library for cloud-scale image processing and computer vision.
When your agent uses it
- Asked to reduce code duplication
- Unify an operators kernels/bindings
- Remove dead code
- Verify a refactor changed nothing observable
Example prompts
- “/refactor-op”
Requirements
- Python 3
Workflow steps
3 steps, taken from the first numbered list in SKILL.md.
- assess — python3 tools/refactor_op.py (scope with --domain impl|api|xcut;
- apply (only if asked to fix) — apply the finding's named corrective action on the
- verify — when the MR scope includes a refactor, run
What it can do on your machine
Read from SKILL.md and the folder at commit b051f32. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
python3
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Refactor Op loads about 1.5k tokens when it runs. Until then it costs about 125 tokens; SKILL.md has 640 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
Its licence (Custom licence) doesn't allow us to republish the file, so here is its outline and opening line. It has 640 words (~1,455 tokens).
“[//]: # "SPDX-FileCopyrightText: Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved." [//]: # "SPDX-License-Identifier: Apache-2.0"”
- name
- refactor-op
Files
Just SKILL.md in .agents/skills/refactor-op of CVCUDA/CV-CUDA.
Open the folder on GitHubat commit b051f32
Compare with similar skills
Refactor Op next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Refactor Op this skillCVCUDA/CV-CUDA | 2.7k | — | ~1.5k | Automated safety check: Pass | Custom licence | |
| Code RefinerMathews-Tom/armory | 328 | — | ~3.1k | Automated safety check: Pass | MIT | |
| Cutlass SkillslowlyC/agent-gpu-skills | 169 | — | ~1.3k | Automated safety check: Pass | MIT | |
| Triton SkillslowlyC/agent-gpu-skills | 169 | — | ~1.3k | Automated safety check: Pass | MIT | |
| Vllm Deploy Simplevllm-project/vllm-skills | 103 | — | ~1.6k | Automated safety check: Pass | Apache-2.0 | |
| Hyperpod Version Checkerawslabs/agent-plugins | 915 | 1 repos | ~910 | Automated safety check: Pass | Apache-2.0 |
Similar skills
Cutlass Skill
slowlyC/agent-gpu-skills
Write, debug, and optimize CUTLASS, CuTe, and CuTeDSL GPU kernels from local upstream source, examples, and headers.
Triton Skill
slowlyC/agent-gpu-skills
Write, debug, and optimize Triton and Gluon GPU kernels from local upstream tutorials, production kernels, language definitions, and compiler source.
Vllm Deploy Simple
vllm-project/vllm-skills
Quick install and deploy vLLM, start serving with a simple LLM, and test OpenAI API.
- Official
Hyperpod Version Checker
awslabs/agent-plugins
Check and compare software component versions on SageMaker HyperPod cluster nodes - NVIDIA drivers, CUDA toolkit, cuDNN, NCCL, EFA, AWS OFI NCCL, GDRCopy, MPI, Neuron SDK (Trainium/Inferentia)…
Quark Env Preflight
amd/Quark
Collect and normalize environment facts (OS, Python, GPU, CUDA/ROCm, container state) before Quark installation or PTQ planning.
More from CVCUDA/CV-CUDA
All 12 skills in this repoOptimize Op
CVCUDA/CV-CUDA
Drive a single-operator optimization campaign per .agents/guidance/OPTIMIZATIONGUIDELINES.md, with a deterministically enforced definition-of-done and versioned MR summary.
Review Op
CVCUDA/CV-CUDA
Review a CV-CUDA operator end-to-end (support / test / bench / docs coverage).
Optimize Op Verify
CVCUDA/CV-CUDA
Verify a CV-CUDA optimization campaign's deterministic definition-of-done and concise versioned MR summary per .agents/guidance/OPTIMIZATIONGUIDELINES.md.
Make Op
CVCUDA/CV-CUDA
Add a new CV-CUDA operator end-to-end per .agents/guidance/MAKEOPGUIDELINES.md, with a deterministically-enforced definition-of-done.
Make Op Scaffold
CVCUDA/CV-CUDA
Scaffold a new CV-CUDA operator — a complete, wired, building skeleton — and delegate the implementation to a human or another AI.
Make Op Verify
CVCUDA/CV-CUDA
Verify a new CV-CUDA operator against the deterministic final regression checklist (the /make-op done-gate).
Related
Works with
Categories
Questions about Refactor Op
What does Refactor Op do?
Find and safely apply per-operator refactoring / redundancy-reduction opportunities in a CV-CUDA operator (near-duplicate Tensor/VarShape kernels, reinvented shared utilities, dead code). Refactor Op is an agent skill from CVCUDA/CV-CUDA. Find and safely apply per-operator refactoring / redundancy-reduction opportunities in a CV-CUDA operator (near-duplicate Tensor/VarShape kernels, reinvented shared utilities, dead code).
When should I use Refactor Op?
Refactor Op fits situations like: asked to reduce code duplication; unify an operators kernels/bindings; remove dead code; verify a refactor changed nothing observable.
How do I install Refactor Op in Claude Code?
Run `npx skills add CVCUDA/CV-CUDA --skill refactor-op -a claude-code`. Or copy the skill folder (.agents/skills/refactor-op in CVCUDA/CV-CUDA) into .claude/skills/refactor-op in your project. Claude Code loads it when a task matches its description.
How do I install Refactor Op in Codex?
Run `npx skills add CVCUDA/CV-CUDA --skill refactor-op -a codex`. Or copy the skill folder (.agents/skills/refactor-op in CVCUDA/CV-CUDA) into .agents/skills/refactor-op in your project. Codex loads it when a task matches its description.
Can I use Refactor Op in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add CVCUDA/CV-CUDA --skill refactor-op -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/refactor-op, .gemini/skills/refactor-op, .github/skills/refactor-op and .opencode/skills/refactor-op in your project.
What does Refactor Op need to run?
Going by SKILL.md and its folder, Refactor Op needs the command-line tools its instructions call (python3). Our summary lists: Python 3.
Does Refactor Op access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Refactor Op safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Refactor Op use?
Refactor Op has a licence file (the repository's licence) that doesn't match a standard licence. Read it on GitHub before reusing the skill.
How many tokens does Refactor Op use?
About 1.5k tokens (SKILL.md is roughly 5.8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Refactor Op?
Skills that share tags, products or a category with Refactor Op: Code Refiner (Mathews-Tom/armory, 328 stars), Cutlass Skill (slowlyC/agent-gpu-skills, 169 stars), Triton Skill (slowlyC/agent-gpu-skills, 169 stars) and Vllm Deploy Simple (vllm-project/vllm-skills, 103 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Refactor Op?
CVCUDA (a GitHub organization) maintains it in CVCUDA/CV-CUDA, which has 2,728 GitHub stars. The repository holds 12 skills in this directory. The repository was last updated on September 16, 2026.
Source: CVCUDA/CV-CUDA on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.