Quark Torch Quant Perf
amd/Quark
Run, resume, monitor, diagnose, and report Quark Quant-Perf workflows for PyTorch and HuggingFace transformers models.
A skill your agent uses when planning or auditing TorchVision reference training/evaluation workflows for classification, quantization, detection, segmentation, video classification, optical flow…
Install Training References
$ npx skills add VectorSpaceLab/AREX-Skill --skill training-references -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install VectorSpaceLab/AREX-Skill training-references --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/VectorSpaceLab/AREX-Skill.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/repositories/repo-skills/torchvision/sub-skills/training-references .claude/skills/training-references && 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 "training-references" agent skill from https://github.com/VectorSpaceLab/AREX-Skill/tree/main/skills/repositories/repo-skills/torchvision/sub-skills/training-references into .claude/skills/training-references/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "training-references", 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/VectorSpaceLab/AREX-Skill/tree/main/skills/repositories/repo-skills/torchvision/sub-skills/training-referencesType 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 VectorSpaceLab/AREX-Skill --skill training-references -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install VectorSpaceLab/AREX-Skill training-references --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/VectorSpaceLab/AREX-Skill.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/repositories/repo-skills/torchvision/sub-skills/training-references .agents/skills/training-references && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "training-references" agent skill from https://github.com/VectorSpaceLab/AREX-Skill/tree/main/skills/repositories/repo-skills/torchvision/sub-skills/training-references into .agents/skills/training-references/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "training-references", 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 VectorSpaceLab/AREX-Skill --skill training-references -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install VectorSpaceLab/AREX-Skill training-references --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/VectorSpaceLab/AREX-Skill.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/repositories/repo-skills/torchvision/sub-skills/training-references .cursor/skills/training-references && 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 "training-references" agent skill from https://github.com/VectorSpaceLab/AREX-Skill/tree/main/skills/repositories/repo-skills/torchvision/sub-skills/training-references into .cursor/skills/training-references/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "training-references", 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/VectorSpaceLab/AREX-Skill.git --path skills/repositories/repo-skills/torchvision/sub-skills/training-references--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 VectorSpaceLab/AREX-Skill --skill training-references -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install VectorSpaceLab/AREX-Skill training-references --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/VectorSpaceLab/AREX-Skill.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/repositories/repo-skills/torchvision/sub-skills/training-references .gemini/skills/training-references && 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 "training-references" agent skill from https://github.com/VectorSpaceLab/AREX-Skill/tree/main/skills/repositories/repo-skills/torchvision/sub-skills/training-references into .gemini/skills/training-references/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "training-references", 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 VectorSpaceLab/AREX-Skill training-referencesInstalls 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 VectorSpaceLab/AREX-Skill --skill training-references -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/VectorSpaceLab/AREX-Skill.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/repositories/repo-skills/torchvision/sub-skills/training-references .github/skills/training-references && 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 "training-references" agent skill from https://github.com/VectorSpaceLab/AREX-Skill/tree/main/skills/repositories/repo-skills/torchvision/sub-skills/training-references into .github/skills/training-references/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "training-references", 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 VectorSpaceLab/AREX-Skill --skill training-references -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install VectorSpaceLab/AREX-Skill training-references --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/VectorSpaceLab/AREX-Skill.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/repositories/repo-skills/torchvision/sub-skills/training-references .opencode/skills/training-references && 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 "training-references" agent skill from https://github.com/VectorSpaceLab/AREX-Skill/tree/main/skills/repositories/repo-skills/torchvision/sub-skills/training-references into .opencode/skills/training-references/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "training-references", 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
training-references- GitHub stars
- 328
- Token cost
- ~684 tokens
- SKILL.md length
- 259 words
- Files
- 5 (incl. scripts, references)
- Skills in repo
- 157
- Repo updated
- First seen
- Licence
- BSD-3-Clause
At a glance
A skill your agent uses when planning or auditing TorchVision reference training/evaluation workflows for classification, quantization, detection, segmentation, video classification, optical flow…
- Works in 5 steps: Identify the task family:… → Read references/task-command-recipes.md… → Check… → …
- Auditing TorchVision reference training/evaluation workflows for classification
- SKILL.md covers Route first, Safe workflow, Safety labels and Bundled helper
- Runs Python scripts from its folder; calls python
What it does
Training References is an agent skill from VectorSpaceLab/AREX-Skill. Use when planning or auditing TorchVision reference training/evaluation workflows for classification, quantization, detection, segmentation, video classification, optical flow, similarity learning, or stereo depth without launching expensive jobs.
Its SKILL.md is about 680 tokens, which your agent loads only when the skill is triggered. The skill folder holds 6 other files, including scripts and reference files (for example `references/data-layouts-and-presets.md`, `references/task-command-recipes.md` and `references/troubleshooting.md`).
It sits in AI & LLM Engineering, covering LLM inference and serving. It works with PyTorch. The repository describes itself as: A Skill Library for Automated Machine Learning. The licence is BSD-3-Clause.
When your agent uses it
- Auditing TorchVision reference training/evaluation workflows for classification
- Video classification
- Similarity learning
- Stereo depth without launching expensive jobs
Example prompts
- “/training-references”
Requirements
- Python 3
Workflow steps
5 steps, taken from the first numbered list in SKILL.md.
- Identify the task family: classification, quantization, detection, segmentation, video classification, optical flow, similarity learning…
- Read references/task-command-recipes.md for concrete command skeletons and safety labels.
- Check references/data-layouts-and-presets.md for expected dataset layout, preset, and preprocessing assumptions.
- Use scripts/inspect_reference_args.py --list or --task to inspect known argument families without importing or running training code.
- Read references/troubleshooting.md before advising a user to run any command that needs datasets, GPUs, distributed launch, checkpoints…
What it can do on your machine
Read from SKILL.md and the folder at commit ac3fe1a. 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
Ships 1 file in scripts/ (Python), which the agent can run.
Shell commands in SKILL.md call:
python
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
Training References loads about 684 tokens when it runs, and up to ~4.7k if it reads all its reference files. Until then it costs about 67 tokens; SKILL.md has 259 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); the scripts in this folder are not scanned.
SKILL.md
The full file from VectorSpaceLab/AREX-Skill at commit ac3fe1a, republished under its BSD-3-Clause licence (© VectorSpaceLab). 259 words, ~684 tokens.
Download SKILL.mdSave it as
.claude/skills/training-references/SKILL.md (or your agent's skills folder). This skill also uses 4 other files; get the full folder from GitHub.- name
- training-references
- description
- Use when planning or auditing TorchVision reference training/evaluation workflows for classification, quantization, detection, segmentation, video classification, optical flow, similarity learning, or stereo depth without launching expensive jobs.
- disable-model-invocation
- true
- metadata.disco-role
- operating
- license
- BSD 3-Clause
TorchVision Training References
Use this sub-skill to turn TorchVision's official reference scripts into safe command plans, dataset-layout checks, and troubleshooting notes. The reference scripts are training baselines rather than stable package APIs; always treat generated commands as plans to review before running.
Route first
- For model constructors, weight enums, or
weights.transforms()usage, route to../models-and-weights/. - For transform implementation details, TVTensors, masks, boxes, videos, or custom v2 pipelines, route to
../transforms-and-tv-tensors/. - For dataset constructors, downloads, codecs, and tiny fixtures, route to
../datasets-io-utils/. - For box utilities, NMS, ROI ops, and detection postprocessing internals, route to
../ops-and-detection/.
Safe workflow
- Identify the task family: classification, quantization, detection, segmentation, video classification, optical flow, similarity learning, or stereo depth.
- Read
references/task-command-recipes.mdfor concrete command skeletons and safety labels. - Check
references/data-layouts-and-presets.mdfor expected dataset layout, preset, and preprocessing assumptions. - Use
scripts/inspect_reference_args.py --listor--task <name>to inspect known argument families without importing or running training code. - Read
references/troubleshooting.mdbefore advising a user to run any command that needs datasets, GPUs, distributed launch, checkpoints, or weight downloads.
Safety labels
- Safe: listing arguments, producing command plans, and reviewing flags.
- Review required: single-process evaluation on already-prepared local data, especially when it may download weights.
- Unsafe by default: full training, distributed
torchrun, dataset downloads, model-url download checks, release scripts, and benchmarks.
Bundled helper
Run the helper from this sub-skill directory or provide its path explicitly:
python scripts/inspect_reference_args.py --list
python scripts/inspect_reference_args.py --task detection
python scripts/inspect_reference_args.py --task classification --format shellThe helper is a static summary adapted from the reference parsers. It does not import TorchVision, import the original scripts, read datasets, download weights, launch distributed jobs, or run training.
© VectorSpaceLab, BSD-3-Clause. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
SKILL.md and 4 other files (scripts, references) in skills/repositories/repo-skills/torchvision/sub-skills/training-references of VectorSpaceLab/AREX-Skill.
- SKILL.md
- references/data-layouts-and-presets.md
- references/task-command-recipes.md
- references/troubleshooting.md
- scripts/inspect_reference_args.py
Open the folder on GitHubat commit ac3fe1a
Compare with similar skills
Training References 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 |
|---|---|---|---|---|---|---|
| Training References this skillVectorSpaceLab/AREX-Skill | 328 | — | ~684 | Automated safety check: Pass | BSD-3-Clause | |
| Quark Torch Quant Perfamd/Quark | 181 | — | ~3k | Automated safety check: Pass | MIT | |
| RWKV Architecture GuideOrchestra-Research/AI-Research-SKILLs | 13k | 3 repos | ~1.8k | Automated safety check: Pass | MIT | |
| ML Engineerdavila7/claude-code-templates | 32k | 10 repos | ~2.3k | Automated safety check: Pass | MIT | |
| Databricks ML Trainingdatabricks/databricks-agent-skills | 345 | — | ~4.6k | Automated safety check: Pass | Custom licence | |
| Quark Torch Routeramd/Quark | 181 | — | ~1.9k | Automated safety check: Pass | MIT |
Similar skills
RWKV Architecture Guide
Orchestra-Research/AI-Research-SKILLs
Explains RWKV, a hybrid that trains in parallel like a GPT and runs inference like an RNN with constant memory per token, plus usage, fine-tuning and troubleshooting.
ML Engineer
davila7/claude-code-templates
Build production ML systems with PyTorch 2.x, TensorFlow, and modern ML frameworks.
- Official
Databricks ML Training
databricks/databricks-agent-skills
Train ML models on Databricks. An agent skill from databricks/databricks-agent-skills.
Quark Torch Router
amd/Quark
Route Quark user goals to the correct atomic skill or workflow.
Quark Torch Router
amd/Quark
Route Quark user goals to the correct atomic skill or workflow.
More from VectorSpaceLab/AREX-Skill
All 157 skills in this repoAgent Lightning
VectorSpaceLab/AREX-Skill
Use this repo skill for Agent Lightning package tasks: authoring trainable agents, tracing rewards and spans, running LightningStore/Trainer loops, using agl CLI services, choosing examples, and…
Agent Tools
VectorSpaceLab/AREX-Skill
A skill your agent uses when configuring LiteLLM for MCP tools, A2A agents, Claude Code/Cursor agent gateway traffic, MCP auth/OAuth, tool permissions, semantic filtering, or agent-specific proxy…
Agents And Awel
VectorSpaceLab/AREX-Skill
Build and debug DB-GPT agents, tools, skills, teams, and AWEL workflows, including deterministic local DAG runs and HTTP-trigger topology without assuming an LLM, credential, or external service.
Agents And Middleware
VectorSpaceLab/AREX-Skill
Work on the actively maintained LangChain v1 agent package: initchatmodel, createagent, structured output, tools, middleware, embeddings initialization, provider routing, and agent runtime…
Agents Workflows
VectorSpaceLab/AREX-Skill
A skill your agent uses for giskard.agents async chat workflows, tools, prompt templates, structured outputs, retries, rate limiting, embeddings, and optional LiteLLM backend.
Alphafold3
VectorSpaceLab/AREX-Skill
A skill your agent uses for AlphaFold 3 input preparation, prediction command planning, output interpretation, and Python API inspection.
Related
Works with
Categories
Questions about Training References
What does Training References do?
A skill your agent uses when planning or auditing TorchVision reference training/evaluation workflows for classification, quantization, detection, segmentation, video classification, optical flow…. Training References is an agent skill from VectorSpaceLab/AREX-Skill. Use when planning or auditing TorchVision reference training/evaluation workflows for classification, quantization, detection, segmentation, video classification, optical flow, similarity learning, or stereo depth without launching expensive jobs.
When should I use Training References?
Training References fits situations like: auditing TorchVision reference training/evaluation workflows for classification; video classification; similarity learning; stereo depth without launching expensive jobs.
How do I install Training References in Claude Code?
Run `npx skills add VectorSpaceLab/AREX-Skill --skill training-references -a claude-code`. Or copy the skill folder (skills/repositories/repo-skills/torchvision/sub-skills/training-references in VectorSpaceLab/AREX-Skill) into .claude/skills/training-references in your project. Claude Code loads it when a task matches its description.
How do I install Training References in Codex?
Run `npx skills add VectorSpaceLab/AREX-Skill --skill training-references -a codex`. Or copy the skill folder (skills/repositories/repo-skills/torchvision/sub-skills/training-references in VectorSpaceLab/AREX-Skill) into .agents/skills/training-references in your project. Codex loads it when a task matches its description.
Can I use Training References 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 VectorSpaceLab/AREX-Skill --skill training-references -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/training-references, .gemini/skills/training-references, .github/skills/training-references and .opencode/skills/training-references in your project.
What does Training References need to run?
Going by SKILL.md and its folder, Training References needs Python for the scripts in its folder and the command-line tools its instructions call (python). Our summary lists: Python 3.
Does Training References 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 Training References 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. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.
What licence does Training References use?
Training References is published under the BSD-3-Clause licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Training References use?
About 684 tokens (SKILL.md is roughly 2.7k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 4.1k tokens, read only when the agent opens those files.
What are the alternatives to Training References?
Skills that share tags, products or a category with Training References: Quark Torch Quant Perf (amd/Quark, 181 stars), RWKV Architecture Guide (Orchestra-Research/AI-Research-SKILLs, 13k stars), ML Engineer (davila7/claude-code-templates, 32k stars) and Databricks ML Training (databricks/databricks-agent-skills, 345 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Training References?
VectorSpaceLab (a GitHub organization) maintains it in VectorSpaceLab/AREX-Skill, which has 328 GitHub stars. The repository holds 157 skills in this directory. The repository was last updated on September 3, 2026.
Source: VectorSpaceLab/AREX-Skill on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.