alterlab-scgpt

alterlab-scgpt is a skill for Claude Code from AlterLab-IEU/AlterLab-Academic-Skills. It costs 163 tokens per session (984 once invoked), scanned A, original, MIT.

A tool based on scGPT, a pretrained model for single-cell biology that has learned patterns from large collections of cells. It works with AnnData files, a common format for storing single-cell measurements and their metadata.

In plain words
What is it for?
Use it to annotate cell types, create cell or gene embeddings, integrate batches, and transfer knowledge to new single-cell datasets.
Why use it?
It can help label and compare cells in a new dataset without always training a model from the beginning, while also providing numerical representations of cells and genes.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter.

Part of the alterlab-bioinformatics plugin — 38 skills shipped together

Good fit Use it to annotate cell types, create cell or gene embeddings, integrate batches, and transfer knowledge to new single-cell datasets.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/alterlab-ieu/alterlab-academic-skills/alterlab-scgpt
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 AlterLab-IEU/AlterLab-Academic-Skills --skill alterlab-scgpt
Clone the repo
git clone --depth 1 https://github.com/AlterLab-IEU/AlterLab-Academic-Skills

Made for: Claude Code.

Or install alterlab-bioinformatics, the plugin that ships this one along with the rest of its 38 skills.

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 alterlab-scgpt

README.md
[![agentmods](https://agentmods.dev/badge/skills/alterlab-ieu/alterlab-academic-skills/alterlab-scgpt/github.svg)](https://agentmods.dev/skills/alterlab-ieu/alterlab-academic-skills/alterlab-scgpt)
Your own site
<a href="https://agentmods.dev/skills/alterlab-ieu/alterlab-academic-skills/alterlab-scgpt"><img src="https://agentmods.dev/badge/skills/alterlab-ieu/alterlab-academic-skills/alterlab-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 alterlab-scgpt

Your own site · 80×15
<a href="https://agentmods.dev/skills/alterlab-ieu/alterlab-academic-skills/alterlab-scgpt"><img src="https://agentmods.dev/badge/skills/alterlab-ieu/alterlab-academic-skills/alterlab-scgpt.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 163 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 984 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 $0.00163 $0.00984
Opus 5 $0.00081 $0.00492
Sonnet 5 $0.00033 $0.00197
Haiku 4.5 $0.00016 $0.00098

Measured 12d ago against content hash ed276271b0f9, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

alterlab-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 12d 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/bioinformatics/alterlab-scgpt/SKILL.md · 81 lines

How it starts

The opening of the file, as written. The whole thing — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.

scGPT (single-cell foundation model)

Overview

scGPT (Cui et al., Nature Methods 2024; bowang-lab/scGPT) is a transformer foundation model pretrained on tens of millions of cells. It provides zero-shot and fine-tuned cell-type annotation, gene and cell embeddings, batch integration, and gene-regulatory / perturbation inference — all operating on AnnData (.h5ad) objects.

Its niche vs. the existing single-cell skills: scGPT is the pretrained-transformer route. For probabilistic latent-variable models use alterlab-scvi-tools; for the conventional Scanpy analysis pipeline use alterlab-scanpy; scGPT complements both.

When to Use This Skill

Use this skill when the user wants to:

  • Annotate cell types with a pretrained foundation model (zero-shot or fine-tuned).
  • Generate scGPT embeddings for cells or genes.
  • Integrate batches using the transformer's representation.
  • Run zero-shot inference / transfer to a new dataset without training from scratch.

Does NOT Trigger

Scenario Use instead
Probabilistic integration / latent model (scVI, scANVI) alterlab-scvi-tools
Standard QC → cluster → UMAP → differential expression alterlab-scanpy
Read/write/wrangle the .h5ad data structure itself alterlab-anndata
RNA velocity alterlab-scvelo
Protein (not single-cell) language models alterlab-esm

Core Capabilities

1. Zero-shot cell embedding & annotation

# bowang-lab/scGPT — API sketch; TODO(verify) against installed scgpt
import scanpy as sc
adata = sc.read_h5ad("cells.h5ad")
# Load a pretrained scGPT checkpoint, embed cells, map to reference cell types.
# (see references/scgpt_usage.md for the exact embed/annotate calls)

Zero-shot mode maps a new dataset onto scGPT's learned space without training — fast triage of cell identities. Fine-tuning on labeled reference data improves accuracy on a specific tissue.

2. Embeddings for downstream analysis

Read the full file on GitHub · 81 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. 12d ago First seen · 81 lines · 163 tokens per session scan A ed276271b0f9

Subscribe to this mod's changes

alterlab-scgpt is a skill published in the GitHub repository AlterLab-IEU/AlterLab-Academic-Skills (66 stars, last pushed 8d ago), licensed MIT. It adds 163 tokens to every session and 984 once invoked, about $0.0008 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

r-spss-syntax-architect

A guide for turning research hypotheses into repeatable R or SPSS code for statistical analysis. It covers panel data, where the same companies or other units are observed over time, as well as interaction effects, curves, and mediation.

Nero1688/claude-academic-skills · 406 tokens

ob-hrm-scale-adaptor

A guide for adapting organisational behaviour and human-resources survey scales across languages and cultures. It covers permission checks, translation and back-translation, expert review, participant interviews, and tests of whether groups interpret the scale comparably.

Nero1688/claude-academic-skills · 318 tokens

thesis-consistency-audit

A consistency audit for quantitative master’s and doctoral theses in management, finance, or strategy. It checks whether numbers, tables, analyses, and claims agree, and can inspect hidden author information in office documents.

Nero1688/claude-academic-skills · 500 tokens

management-figure

A chart-making toolkit for evidence-based management, finance, and strategy research. It creates publication-ready plots from regression results and tracking data, including coefficient, interaction, group-comparison, trend, and curved-relationship charts.

Nero1688/claude-academic-skills · 374 tokens

reproducibility-architect

A guide for packaging research so another person can rerun its data processing and analysis. A replication package is the project files, instructions, code, data guidance, and software details needed to reproduce published results.

Nero1688/claude-academic-skills · 606 tokens

journal-submission-scout

A research tool for choosing a journal for a completed paper. It searches for journals that publish similar work, compares public information such as citation data, fees, open-access listing, and review practices, and screens for warning signs of predatory journals, which charge authors without providing trustworthy publishing services.

Nero1688/claude-academic-skills · 479 tokens