Agent skill

Prepare Dataset

Prepare datasets for training on VAST storage (NFS - VAST rechunked zarr v3 pipeline).

BSD-3-ClauseAuto-check passedAI & LLM Engineering

Install Prepare Dataset

skills CLI

$ npx skills add mehta-lab/VisCy --skill prepare-dataset -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI

$ gh skill install mehta-lab/VisCy prepare-dataset --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy

$ git clone --depth 1 https://github.com/mehta-lab/VisCy.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/prepare-dataset .claude/skills/prepare-dataset && rm -rf skills-src

Use ~/.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/

Paste into Claude Code

Install the "prepare-dataset" agent skill from https://github.com/mehta-lab/VisCy/tree/main/.claude/skills/prepare-dataset into .claude/skills/prepare-dataset/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "prepare-dataset", 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

$skill-installer install https://github.com/mehta-lab/VisCy/tree/main/.claude/skills/prepare-dataset

Type 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.

skills CLI

$ npx skills add mehta-lab/VisCy --skill prepare-dataset -a codex

Project install goes to .agents/skills/; add -g for ~/.codex/skills/.

GitHub CLI

$ gh skill install mehta-lab/VisCy prepare-dataset --agent codex

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/mehta-lab/VisCy.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.claude/skills/prepare-dataset .agents/skills/prepare-dataset && rm -rf skills-src

Use ~/.agents/skills/ instead of .agents/skills for a personal install.

Codex skills documentation · loads skills from .agents/skills/

Paste into Codex

Install the "prepare-dataset" agent skill from https://github.com/mehta-lab/VisCy/tree/main/.claude/skills/prepare-dataset into .agents/skills/prepare-dataset/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "prepare-dataset", 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.

skills CLI

$ npx skills add mehta-lab/VisCy --skill prepare-dataset -a cursor

Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.

GitHub CLI

$ gh skill install mehta-lab/VisCy prepare-dataset --agent cursor

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/mehta-lab/VisCy.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.claude/skills/prepare-dataset .cursor/skills/prepare-dataset && rm -rf skills-src

Use ~/.cursor/skills/ instead of .cursor/skills for a personal install.

Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/

Paste into Cursor

Install the "prepare-dataset" agent skill from https://github.com/mehta-lab/VisCy/tree/main/.claude/skills/prepare-dataset into .cursor/skills/prepare-dataset/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "prepare-dataset", 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

$ gemini skills install https://github.com/mehta-lab/VisCy.git --path .claude/skills/prepare-dataset

--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.

skills CLI

$ npx skills add mehta-lab/VisCy --skill prepare-dataset -a gemini-cli

Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.

GitHub CLI

$ gh skill install mehta-lab/VisCy prepare-dataset --agent gemini-cli

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/mehta-lab/VisCy.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.claude/skills/prepare-dataset .gemini/skills/prepare-dataset && rm -rf skills-src

Use ~/.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/

Paste into Gemini CLI

Install the "prepare-dataset" agent skill from https://github.com/mehta-lab/VisCy/tree/main/.claude/skills/prepare-dataset into .gemini/skills/prepare-dataset/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "prepare-dataset", 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.

GitHub CLI

$ gh skill install mehta-lab/VisCy prepare-dataset

Installs 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).

skills CLI

$ npx skills add mehta-lab/VisCy --skill prepare-dataset -a github-copilot

Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.

Manual copy

$ git clone --depth 1 https://github.com/mehta-lab/VisCy.git skills-src && mkdir -p .github/skills && cp -r skills-src/.claude/skills/prepare-dataset .github/skills/prepare-dataset && rm -rf skills-src

Use ~/.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/

Paste into GitHub Copilot

Install the "prepare-dataset" agent skill from https://github.com/mehta-lab/VisCy/tree/main/.claude/skills/prepare-dataset into .github/skills/prepare-dataset/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "prepare-dataset", 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.

skills CLI

$ npx skills add mehta-lab/VisCy --skill prepare-dataset -a opencode

OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.

GitHub CLI

$ gh skill install mehta-lab/VisCy prepare-dataset --agent opencode

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/mehta-lab/VisCy.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.claude/skills/prepare-dataset .opencode/skills/prepare-dataset && rm -rf skills-src

Use ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.

OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/

Paste into OpenCode

Install the "prepare-dataset" agent skill from https://github.com/mehta-lab/VisCy/tree/main/.claude/skills/prepare-dataset into .opencode/skills/prepare-dataset/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "prepare-dataset", 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: prepare-dataset
GitHub stars: 104
Token cost: ~1.1k tokens
SKILL.md length: 278 words
Files: 1
Skills in repo: 4
Repo updated: Oct 8, 2026
First seen: Oct 7, 2026
Licence: BSD-3-Clause

At a glance

Prepare datasets for training on VAST storage (NFS - VAST rechunked zarr v3 pipeline).

Works in 5 steps: Airtable validation — dataset must be… → Discover wells/channels — reads NFS zarr… → biahub concatenate — rechunks to zarr v3… → …
The user asks to prepare a dataset
SKILL.md covers Overview, Pipeline Steps, Key Files and Commands, plus 5 more sections
Calls uv and conda

What it does

Prepare Dataset is an agent skill from mehta-lab/VisCy. Prepare datasets for training on VAST storage (NFS - VAST rechunked zarr v3 pipeline). Use when the user asks to "prepare a dataset", "run prepare", "rechunk dataset", "copy dataset to VAST", "run QC and preprocess", or references the prepare CLI.

Its SKILL.md is about 1.1k 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 AI & LLM Engineering. It works with Zarr and Airtable. The repository describes itself as: computer vision models for single-cell phenotyping. The licence is BSD-3-Clause.

When your agent uses it

The user asks to prepare a dataset
Rechunk dataset
Copy dataset to VAST
Run QC and preprocess

Example prompts

“prepare a dataset”
“run prepare”
“rechunk dataset”
“/prepare-dataset”

Requirements

Python 3

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

Airtable validation — dataset must be registered
Discover wells/channels — reads NFS zarr via iohub; auto-detects raw channels (Phase3D + raw *)
biahub concatenate — rechunks to zarr v3 with sharding (submits own SLURM jobs via submitit)
Copy tracking zarr — rsync from NFS
QC + preprocess — SLURM job running focus slice QC (GPU) and normalization (CPU) in parallel

What it can do on your machine

Read from SKILL.md and the folder at commit 4b62365. 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:
- uv
- conda
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md. Its commands use uv, which can reach the network depending on how they are called.
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

Prepare Dataset loads about 1.1k tokens when it runs. Until then it costs about 67 tokens; SKILL.md has 278 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it: ~67
When it runs · the whole SKILL.md, loaded when a task matches: ~1.1k

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

The full file from mehta-lab/VisCy at commit 4b62365, republished under its BSD-3-Clause licence (© mehta-lab). 278 words, ~1,133 tokens.

Download SKILL.mdSave it as .claude/skills/prepare-dataset/SKILL.md (or your agent's skills folder).

name: prepare-dataset
description: Prepare datasets for training on VAST storage (NFS -> VAST rechunked zarr v3 pipeline). Use when the user asks to "prepare a dataset", "run prepare", "rechunk dataset", "copy dataset to VAST", "run QC and preprocess", or references the `prepare` CLI.

Prepare Dataset for Training (NFS -> VAST)

Overview

This skill runs the prepare CLI from applications/airtable/ to create rechunked zarr v3 copies of NFS datasets on VAST storage, with QC (focus slice) and preprocessing (normalization stats).

Pipeline Steps

Airtable validation — dataset must be registered
Discover wells/channels — reads NFS zarr via iohub; auto-detects raw channels (Phase3D + raw *)
biahub concatenate — rechunks to zarr v3 with sharding (submits own SLURM jobs via submitit)
Copy tracking zarr — rsync from NFS
QC + preprocess — SLURM job running focus slice QC (GPU) and normalization (CPU) in parallel

Key Files

CLI: applications/airtable/src/airtable_utils/prepare_cli.py
Core logic: applications/airtable/src/airtable_utils/prepare.py
Default config: applications/airtable/configs/prepare_config.yml

Commands

bash

# Check status of one or more datasets
uv run --package airtable-utils \
    prepare status <dataset_name> [<dataset_name> ...] \
    -c applications/airtable/configs/prepare_config.yml

# Dry run (generate configs + scripts, don't execute)
uv run --package airtable-utils \
    prepare run <dataset_name> \
    -c applications/airtable/configs/prepare_config.yml --dry-run

# Full run
uv run --package airtable-utils \
    prepare run <dataset_name> \
    -c applications/airtable/configs/prepare_config.yml

# Force overwrite existing VAST zarr
uv run --package airtable-utils \
    prepare run <dataset_name> \
    -c applications/airtable/configs/prepare_config.yml --force

Running Multiple Datasets

Run prepare run sequentially for each dataset. The concatenation step blocks until biahub's internal SLURM jobs complete, then submits the QC+preprocess SLURM job. Example:

bash

for ds in 2025_01_28_A549_G3BP1_ZIKV_DENV 2025_04_15_A549_H2B_CAAX_ZIKV_DENV; do
    uv run --package airtable-utils \
        prepare run "$ds" \
        -c applications/airtable/configs/prepare_config.yml
done

Output Layout

/hpc/projects/organelle_phenotyping/datasets/{dataset_name}/
  {dataset_name}.zarr       # zarr v3 rechunked (OME-Zarr 0.5)
  tracking.zarr             # copied from NFS
  crop_concat.yml           # generated biahub config
  qc_config.yml             # generated QC config
  sbatch_overrides.sh       # SLURM overrides for biahub (if configured)
  01_concatenate.sh         # bash: biahub concatenate + tracking copy
  02_qc_preprocess.sh       # SLURM: QC + preprocess

Config Reference

The config at applications/airtable/configs/prepare_config.yml has these key settings:

Section	Field	Default	Notes
`concatenate`	`channel_names`	`null` (auto-detect)	Set explicitly to override; auto picks `Phase3D` + `raw *`
`concatenate`	`chunks_czyx`	`[1, 16, 256, 256]`	~4MB chunks for training
`concatenate`	`shards_ratio`	`[1, 1, 8, 8, 8]`	Sharding for zarr v3
`concatenate`	`sbatch_overrides`	`{partition: preempted}`	Overrides biahub's internal SLURM via `-sb`
`qc`	`channel_names`	`[Phase3D]`	Channels for focus slice detection
`slurm.qc_preprocess`	`partition`	`gpu`	QC needs GPU for torch FFT
`slurm.qc_preprocess`	`cpus_per_task`	`16`
`slurm.qc_preprocess`	`time`	`01:00:00`

Extracting Dataset Names from a Collection YAML

To get unique dataset names from a collection:

bash

uv run python3 -c "
import yaml
from pathlib import Path
with open('path/to/collection.yml') as f:
    col = yaml.safe_load(f)
datasets = sorted(set(
    Path(e['data_path']).parts[
        Path(e['data_path']).parts.index('organelle_dynamics') + 1
    ]
    for e in col['experiments']
    if 'organelle_dynamics' in e['data_path']
))
for d in datasets:
    print(d)
"

Troubleshooting

"Dataset not found in Airtable": Register it first with the airtable-register skill
Channel validation fails: Check channel_names in config; set to null for auto-detection
biahub concatenate fails: Check conda env exists (conda run -n biahub which biahub)
QC/preprocess SLURM job pending: Check squeue -u $USER and partition availability

Files

Just SKILL.md in .claude/skills/prepare-dataset of mehta-lab/VisCy.

Open the folder on GitHubat commit 4b62365

Compare with similar skills

Prepare Dataset 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.

Prepare Dataset compared with similar skills
Skill	Stars	Used in	Tokens	Auto-check	Licence	Repo updated
Prepare Dataset this skillmehta-lab/VisCy	104	—	~1.1k	Automated safety check: Pass	BSD-3-Clause	yesterday
Geotesseraucam-eo/geotessera	354	—	~1.6k	Automated safety check: Pass	MIT	4 days ago
Chroma Vector DatabaseOrchestra-Research/AI-Research-SKILLs	13k	8 repos	~2.3k	Automated safety check: Pass	MIT	3 mo ago
Codebase Managementgiancarloerra/SocratiCode	3.3k	1 repos	~1.8k	Automated safety check: Pass	AGPL-3.0	yesterday
Pgvector Semantic Searchtimescale/pg-aiguide	1.9k	1 repos	~3.8k	Automated safety check: Pass	Apache-2.0	yesterday
Hermes Memory Providersmnemosyne-oss/mnemosyne	3.4k	—	~1.8k	Automated safety check: Pass	MIT	5 days ago

Similar skills

Geotessera
ucam-eo/geotessera
Read Tessera satellite embeddings with the geotessera Python library.
354 GitHub stars~1.6k tokensUpdated 4 days ago
AI & LLM EngineeringAuto-check passed
Chroma Vector Database
Orchestra-Research/AI-Research-SKILLs
Shows how to store documents and embeddings in Chroma, query them by similarity with metadata filters, and persist them to disk for RAG and semantic search projects.
13k GitHub starsUsed in 8 repos~2.3k tokens
AI & LLM EngineeringAuto-check passed
Codebase Management
giancarloerra/SocratiCode
Set up, index, and manage SocratiCode codebase indexing. An agent skill from giancarloerra/SocratiCode.
3.3k GitHub starsUsed in 1 repo~1.8k tokens
AI & LLM EngineeringAuto-check passed
Pgvector Semantic Search
timescale/pg-aiguide
A skill your agent uses for setting up vector similarity search with pgvector for AI/ML embeddings, RAG applications, or semantic search.
1.9k GitHub starsUsed in 1 repo~3.8k tokens
AI & LLM EngineeringAuto-check passed
Hermes Memory Providers
mnemosyne-oss/mnemosyne
Install and configure Mnemosyne as a Hermes Agent memory provider — local SQLite with vector search, episodic consolidation, and temporal knowledge graphs.
3.4k GitHub stars~1.8k tokensUpdated 5 days ago
AI & LLM EngineeringAuto-check passed
Gptqmodel Tokenizer Normalization
ModelCloud/GPTQModel
Diagnose and correct GPT-QModel tokenizer initialization, tokenization normalization, special-token handling, prompt rendering, and chat-template problems.
1.3k GitHub stars~1.1k tokensUpdated yesterday
AI & LLM EngineeringAuto-check passed

Questions about Prepare Dataset

What does Prepare Dataset do?

Prepare datasets for training on VAST storage (NFS - VAST rechunked zarr v3 pipeline). Prepare Dataset is an agent skill from mehta-lab/VisCy. Prepare datasets for training on VAST storage (NFS - VAST rechunked zarr v3 pipeline).

When should I use Prepare Dataset?

Prepare Dataset fits situations like: the user asks to prepare a dataset; rechunk dataset; copy dataset to VAST; run QC and preprocess.

How do I install Prepare Dataset in Claude Code?

Run `npx skills add mehta-lab/VisCy --skill prepare-dataset -a claude-code`. Or copy the skill folder (.claude/skills/prepare-dataset in mehta-lab/VisCy) into .claude/skills/prepare-dataset in your project. Claude Code loads it when a task matches its description.

How do I install Prepare Dataset in Codex?

Run `npx skills add mehta-lab/VisCy --skill prepare-dataset -a codex`. Or copy the skill folder (.claude/skills/prepare-dataset in mehta-lab/VisCy) into .agents/skills/prepare-dataset in your project. Codex loads it when a task matches its description.

Can I use Prepare Dataset 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 mehta-lab/VisCy --skill prepare-dataset -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/prepare-dataset, .gemini/skills/prepare-dataset, .github/skills/prepare-dataset and .opencode/skills/prepare-dataset in your project.

What does Prepare Dataset need to run?

Going by SKILL.md and its folder, Prepare Dataset needs the command-line tools its instructions call (uv and conda). Our summary lists: Python 3.

Does Prepare Dataset access the network?

SKILL.md contains no URLs. Its commands use uv, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Prepare Dataset 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 Prepare Dataset use?

Prepare Dataset is published under the BSD-3-Clause licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Prepare Dataset use?

About 1.1k tokens (SKILL.md is roughly 4.5k 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 Prepare Dataset?

Skills that share tags, products or a category with Prepare Dataset: Geotessera (ucam-eo/geotessera, 354 stars), Chroma Vector Database (Orchestra-Research/AI-Research-SKILLs, 13k stars), Codebase Management (giancarloerra/SocratiCode, 3.3k stars) and Pgvector Semantic Search (timescale/pg-aiguide, 1.9k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Prepare Dataset?

mehta-lab (a GitHub organization) maintains it in mehta-lab/VisCy, which has 104 GitHub stars. The repository holds 4 skills in this directory. The repository was last updated on October 8, 2026.

Source: mehta-lab/VisCy on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.

Install Prepare Dataset

Facts

At a glance

What it does

When your agent uses it

Example prompts

Requirements

Workflow steps

What it can do on your machine

Tool permissions

Runs code

Network

Credentials

Context cost

Safety

SKILL.md

Prepare Dataset for Training (NFS -> VAST)

Overview

Pipeline Steps

Key Files

Commands

Running Multiple Datasets

Output Layout

Config Reference

Extracting Dataset Names from a Collection YAML

Troubleshooting

Files

Compare with similar skills

Similar skills

More from mehta-lab/VisCy

Related

Questions about Prepare Dataset