single-cell-foundation-model-scgpt

single-cell-foundation-model-scgpt is a skill for Claude Code, Codex from PharMolix/OpenBioMed. It costs 83 tokens per session (808 once invoked), scanned A, original, MIT.

A guide for using the scGPT single-cell foundation model, which learns patterns from gene-expression data in individual cells.

In plain words
What is it for?
Use it for scGPT preprocessing and binning, vocabulary matching, cell embeddings, reference mapping, fine-tuning, and tutorials for annotation, integration, gene-regulatory networks, or perturbation.
Why use it?
It helps prepare data in the format scGPT expects and checks the model vocabulary and checkpoint files before training or inference.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it for scGPT preprocessing and binning, vocabulary matching, cell embeddings, reference mapping, fine-tuning, and tutorials for annotation, integration, gene-regulatory networks, or perturbation.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/pharmolix/openbiomed/single-cell-foundation-model-scrna-seq-scgpt
About the project

OpenBioMed is an agent platform and toolkit collection for biomedical research and drug discovery, covering areas such as molecular design, protein analysis, and single-cell data analysis. It is intended for researchers and provides the biomedical skills listed in the catalogue as workflows for Claude Code.

PharMolix/OpenBioMed · 1,105 stars · on GitHub

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add PharMolix/OpenBioMed --skill single-cell-foundation-model-scrna-seq-scgpt
Clone the repo
git clone --depth 1 https://github.com/PharMolix/OpenBioMed

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for single-cell-foundation-model-scgpt

README.md
[![agentmods](https://agentmods.dev/badge/skills/pharmolix/openbiomed/single-cell-foundation-model-scrna-seq-scgpt/github.svg)](https://agentmods.dev/skills/pharmolix/openbiomed/single-cell-foundation-model-scrna-seq-scgpt)
Your own site
<a href="https://agentmods.dev/skills/pharmolix/openbiomed/single-cell-foundation-model-scrna-seq-scgpt"><img src="https://agentmods.dev/badge/skills/pharmolix/openbiomed/single-cell-foundation-model-scrna-seq-scgpt/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for single-cell-foundation-model-scgpt

Your own site · 80×15
<a href="https://agentmods.dev/skills/pharmolix/openbiomed/single-cell-foundation-model-scrna-seq-scgpt"><img src="https://agentmods.dev/badge/skills/pharmolix/openbiomed/single-cell-foundation-model-scrna-seq-scgpt.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 83 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 808 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00083 $0.00808
Opus 5 $0.00042 $0.00404
Sonnet 5 $0.00017 $0.00162
Haiku 4.5 $0.00008 $0.00081

Measured 13d ago against content hash d94f1e401537, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

single-cell-foundation-model-scgpt scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 13d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/single-cell-foundation-model-scrna-seq-scgpt/SKILL.md · 102 lines

How it starts

The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.

scGPT

Use This Skill When

Use this skill for the local scGPT repository at /DATA/disk0/zhaosy/home/scGPT. It is the right choice when the task involves:

  • preparing AnnData inputs with scGPT's own preprocessing pipeline
  • matching genes to a pretrained scGPT vocabulary
  • tokenizing binned expression inputs for transformer models
  • extracting cell embeddings with pretrained checkpoints
  • fine-tuning scGPT for integration or annotation-style downstream tasks
  • understanding how scGPT expects binned values, special tokens, and batch labels
  • working through scGPT tutorials such as integration, annotation, GRN, perturbation, or reference mapping

Do not use this skill for generic Scanpy work that does not depend on scGPT checkpoints or tokenization.

Start Here

  1. Confirm the checkpoint directory contains args.json, vocab.json, and best_model.pt.
  2. Decide whether the task is fine-tuning, embedding extraction, or tutorial-guided experimentation.
  3. Run preprocessing before tokenization or embedding unless the input has already been prepared for scGPT.
  4. Check vocabulary overlap before spending time on training or inference.

Choose A Path

Preprocess and bin

The core preprocessing path in this repo is scgpt.preprocess.Preprocessor. Typical steps include:

  • filter genes by counts
  • optionally filter cells
  • normalize total counts
  • optionally log1p transform
  • subset highly variable genes
  • bin values into discrete bins and store them in adata.layers["X_binned"]

Fine-tune for integration

The clearest end-to-end example in the local repo is examples/finetune_integration.py. It demonstrates:

  • loading a dataset
  • building str_batch and batch_id
  • preprocessing and HVG selection
  • matching checkpoint vocabulary
  • tokenizing and padding batches
  • training / evaluation for an integration workflow

If the user asks "how should I use scGPT on my AnnData?", this example is often the best starting point.

Extract cell embeddings

Read the full file on GitHub · 102 lines

Files

What ships with it

2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 13d ago First seen · 102 lines · 83 tokens per session scan A d94f1e401537

Subscribe to this mod's changes

single-cell-foundation-model-scgpt is a skill published in the GitHub repository PharMolix/OpenBioMed (1,105 stars, last pushed 1mo ago), licensed MIT. It adds 83 tokens to every session and 808 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

arboreto

Infer gene regulatory networks (GRNs) from gene expression data using scalable algorithms (GRNBoost2, GENIE3). Use when analyzing transcriptomics data (bulk RNA-seq, single-cell RNA-seq) to identify transcription factor-target gene relationships and regulatory interactions. Supports distributed computation for…

K-Dense-AI/scientific-agent-skills · 66 tokens

pyhealth

Build clinical/healthcare deep-learning pipelines with PyHealth — loading EHR/signal/imaging datasets (MIMIC-III/IV, eICU, OMOP, SleepEDF, ChestXray14, EHRShot), defining tasks (mortality, readmission, length-of-stay, drug recommendation, sleep staging, ICD coding, EEG events), instantiating models (Transformer…

K-Dense-AI/scientific-agent-skills · 216 tokens

torchdrug

Build and troubleshoot TorchDrug 0.2.1 workflows for molecular graphs, property prediction, self-supervised pretraining, molecule generation, retrosynthesis, protein representation learning, and knowledge graph reasoning. Use when code imports torchdrug or needs its datasets, models, tasks, or Engine.

K-Dense-AI/scientific-agent-skills · 61 tokens

deepspot-m

Generate transcriptome-wide virtual spatial transcriptomics from H&E histology with DeepSpot-M. Use when you need spatial gene expression in log1p-CPM for 224x224 tiles at about 20x, want to query protein-coding genes by symbol instead of a fixed panel, or want to run prediction across a whole slide after tiling with…

K-Dense-AI/scientific-agent-skills · 80 tokens

nemo-mbridge-perf-expert-parallel-overlap

Validate and use MoE expert-parallel communication overlap in Megatron-Bridge, including overlapmoeexpertparallelcomm, delaywgradcompute, and flex dispatcher backends such as DeepEP and HybridEP.

NVIDIA/skills · 56 tokens

pick-a-pii-model

Select an on-device OpenMed PII model from the committed registry by language, runtime format, and size budget, then require recall validation before deployment. Use when an agent must choose a local PII detector for CPU, Apple Silicon, or a mobile export without relying on live model discovery.

maziyarpanahi/openmed · 64 tokens