BM

biotender-max/awesome-bio-agent-skills

数据科学与机器学习
142 stars 质量 70 趋势 70

A curated collection of AI agent skills for biomedical research, covering genomics, proteomics, single-cell analysis, clinical AI, and protein design.

概览

A curated collection of AI agent skills for biomedical research, covering genomics, proteomics, single-cell analysis, clinical AI, and protein design.

README

Awesome Bio Agent Skills

A curated collection of AI agent skills for biomedical research, covering genomics, proteomics, single-cell analysis, clinical AI, and protein design.

Now tracking NVIDIA BioNeMo Agent Toolkit, Claude Science built-in skills, and the emerging ecosystem of agent-callable tools for life sciences.

1,722 deduplicated skills from 22 open-source repositories, organized into 16 categories. Each skill is a self-contained SKILL.md folder compatible with Claude-based agent frameworks (OpenClaw, NanoClaw, Biomni).


About BioTender

This project is created and maintained by BioTender.

BioTender is a Chinese AI-for-Biology intelligence and media platform covering protein design, structure prediction, virtual cells, AI drug discovery, biological foundation models, scientific agents, AI4Bio startups, and frontier research.


Contents


Genomics

526 skills — WGS/WES analysis, variant annotation, GWAS, CNV, structural variants, haplotype phasing, genome assembly.


Proteomics

167 skills — mass spectrometry analysis, structure prediction, protein design, binding affinity optimization.


Single-Cell Analysis

144 skills — preprocessing, clustering, cell type annotation, trajectory inference, cell communication, multimodal integration.


Biology and AI

236 skills — medical AI, clinical decision support, drug discovery, general biological tools.


Clinical and Medical

152 skills — EHR analysis, clinical trial design, drug interactions, precision medicine, adverse event detection.


Transcriptomics

97 skills — RNA-seq full pipeline, differential expression, alternative splicing, lncRNA, small RNA.


Database Query

63 skills — UniProt, PDB, KEGG, Reactome, GEO, ClinVar, Ensembl, STRING.


Multi-Omics Integration

69 skills — MOFA, DIABLO, single-cell multimodal, spatial transcriptomics.


Bioinformatics Utilities

86 skills — sequence analysis, BLAST, tool chains, pipeline management.


Visualization

48 skills — volcano plots, heatmaps, PCA/UMAP, interactive charts.


Workflow Orchestration

38 skills — Snakemake, Nextflow, CWL, WDL.


Epigenomics

19 skills — ChIP-seq, ATAC-seq, DNA methylation, Hi-C, chromatin state.


Pathway Analysis

15 skills — KEGG, Reactome, GO, GSEA.

Skill Source Description
bio-pathway-enrichment-visualization bioskills Visualize enrichment results using enrichplot package functions. Use when creating publication-quality figures from clusterProfiler results. Covers…
bio-pathway-kegg-pathways bioskills KEGG pathway and module enrichment analysis using clusterProfiler enrichKEGG and enrichMKEGG. Use when identifying metabolic and signaling pathways…
bio-pathway-reactome bioskills Reactome pathway enrichment using ReactomePA package. Use when analyzing gene lists against Reactome’s curated peer-reviewed pathway database. Perf…
bio-pathway-wikipathways bioskills WikiPathways enrichment using clusterProfiler and rWikiPathways. Use when analyzing gene lists against community-curated open-source pathways. Perf…
gsea-enrichment-analysis openclaw Gene set enrichment analysis with correct geneset format handling. Critical guidance for loading pathway databases and running enrichment in OmicVe…
ontology-explorer openclaw Parse, navigate, and query materials science ontology structure (classes, properties, hierarchy). Use when exploring an ontology like CMSO, underst…
ontology-mapper openclaw Map materials science terms, crystal structures, and sample descriptions to ontology classes and properties. Supports any ontology registered in on…
ontology-validator openclaw Validate material sample annotations and data structures against ontology constraints. Use when checking if CMSO annotations are correct, verifying…
tooluniverse-gene-enrichment openclaw Perform comprehensive gene enrichment and pathway analysis using gseapy (ORA and GSEA), PANTHER, STRING, Reactome, and 40+ ToolUniverse tools. Supp…
query-kegg bioclaw Query KEGG for biological pathways and gene info. Use when user asks about metabolic pathways, signaling pathways, pathway genes, or KEGG IDs. Trig…
query-reactome bioclaw Query Reactome for biological pathways and reactions. Use when user asks about signaling cascades, biological processes, pathway diagrams, or react…
generate-image kdense Generate or edit images using AI models (FLUX, Nano Banana 2). Use for general-purpose image generation including photos, illustrations, artwork, v…
scientific-schematics kdense Create publication-quality scientific diagrams using Nano Banana 2 AI with smart iterative refinement. Uses Gemini 3.1 Pro Preview for quality revi…
conn-tool neuroclaw Use this skill whenever the user wants to perform advanced functional connectivity (ROI-to-ROI, seed-to-voxel, ICA) or effective connectivity (PPI,…
Functional Enrichment Analysis (GSEA + ORA) omicsclaw Perform functional enrichment analysis using clusterProfiler on differential expression results with GSEA and ORA.

Metagenomics

9 skills — 16S/ITS amplicon, Kraken2, MetaPhlAn, QIIME2.

Skill Source Description
bio-blast-searches bioskills Run remote BLAST searches against NCBI servers using Biopython Bio.Blast.NCBIWWW. Use when identifying unknown sequences, finding homologs, picking…
bio-crispr-screens-screen-qc bioskills Quality control for pooled CRISPR screens covering library representation, Gini index, log-skew, replicate Pearson and Spearman concordance, essent…
bio-microbiome-qiime2-workflow bioskills QIIME2 command-line workflow for 16S/ITS amplicon analysis. Alternative to DADA2/phyloseq R workflow with built-in provenance tracking. Use when pr…
bio-microbiome-taxonomy-assignment bioskills Taxonomic classification of ASVs using reference databases like SILVA, GTDB, or UNITE. Covers naive Bayes classifiers (DADA2, IDTAXA) and exact mat…
microbiome-cancer-agent openclaw AI-powered analysis of microbiome-cancer interactions including tumor microbiome profiling, immunotherapy response prediction, and microbiome-targe…
meg-skill neuroclaw Use this skill whenever the user wants to process MEG (magnetoencephalography) data including source localization, time-frequency analysis, connect…
wmh-segmentation neuroclaw Use this skill whenever the user wants to perform automated white matter hyperintensity (WMH) segmentation on structural MRI data using the MARS-WM…
bio-annotation omics Functional annotation and taxonomy inference from sequence homology.
tracking-taxonomy-updates omics Track and reconcile taxonomy updates across NCBI, GTDB, ICTV, and community eukaryote frameworks with versioned provenance.

Protein Design

7 skills — RFDiffusion, ProteinMPNN, Boltz, Chai, LigandMPNN.

Skill Source Description
bio-clip-seq-m6a-clip bioskills Map N6-methyladenosine (m6A) RNA modifications at single-nucleotide resolution using miCLIP (Linder 2015), miCLIP2 + m6Aboost machine learning (Kor…
bio-tcr-bcr-analysis-immcantation-analysis bioskills Analyze BCR repertoires for somatic hypermutation, clonal lineages, and B cell phylogenetics using the Immcantation framework. Use when studying B…
mage-antibody-generator openclaw Ab seq forge
ligandmpnn adaptyv Ligand-aware protein sequence design using LigandMPNN. Use this skill when: (1) Designing sequences around small molecules, (2) Enzyme active site…
rfdiffusion adaptyv Generate protein backbones using RFdiffusion, a diffusion-based generative model for de novo protein structure generation. Use this skill when: (1)…
solublempnn adaptyv Solubility-optimized protein sequence design using SolubleMPNN. Use this skill when: (1) Designing for E. coli expression, (2) Optimizing solubilit…
generate_scientific_method_section labclaw Automated SCI-standard Methods section generator from experiment execution records. Parses LabOS skill call chains, structured JSON logs (extract_e…

NVIDIA BioNeMo Skills

17 skills — Official NVIDIA BioNeMo NIM API skills for protein structure prediction, molecular generation, and genomics.

Source: NVIDIA-BioNeMo/bionemo-agent-toolkit · License: Apache-2.0 (code) / CC-BY-4.0 (skills)


Claude Science Skills

29 skills — Reverse-engineered Claude Science built-in skills covering biomolecular modeling, genomics, single-cell analysis, literature synthesis, and scientific compute workflows.

Source: JimLiu/science-skills · License: Apache-2.0 · Reverse-engineered from Claude Science · Synced: fb309c3 (2026-07-01)


Sources

This collection aggregates and deduplicates skills from the following open-source repositories:

Repository Skills Focus
GPTomics/bioSkills 536 Systematic bioinformatics suite from QC to multi-omics.
FreedomIntelligence/OpenClaw-Medical-Skills 359 Medical AI library aggregating 12 specialized sub-repositories.
jaechang-hits/SciAgent-Skills 154 Scientific agent skills for statistics, databases, and clinical decisions.
K-Dense-AI/scientific-agent-skills 102 General scientific computing and HPC workflow skills.
CUHK-AIM-Group/NeuroClaw 86 Neuroimaging skills: sMRI, fMRI, dMRI, EEG with BIDS, FreeSurfer, FSL, fMRIPrep. CUHK AIM Group.
ClawBio/ClawBio 63 Bioinformatics workflow orchestration for GWAS and single-cell.
wu-yc/LabClaw 59 Lab automation and biomedical research skills. Stanford-Princeton AI Co-Scientists.
QSong-github/DrugClaw 57 Drug intelligence skills: DTI, ADR, DDI, pharmacogenomics, repurposing. LangGraph-powered.
ChrisLou-bioinfo/nobel-medicine-minds 55 Cognitive frameworks of 52 Nobel Medicine laureates (2004–2025) as runnable SKILL.md files.
zongtingwei/Bioclaw_Skills_Hub 46 Ten-category biological skills hub.
Runchuan-BU/BioClaw 37 Core bioinformatics tools and database query skills.
fmschulz/omics-skills 29 Single-cell and spatial omics specialized skills.
TianGzlab/OmicsClaw 28 6-omics domain skills: spatial, scRNA-seq, bulk RNA-seq, genomics, proteomics, metabolomics.
adaptyvbio/protein-design-skills 21 Full protein design toolkit: RFDiffusion, ProteinMPNN, Boltz, Chai.
aristoteleo/PantheonOS 18 Single-cell and spatial transcriptomics skills. Dynamo/Spateo team.
EvoScientist/EvoSkills 13 Research-lifecycle skills: ideation, paper planning, experiment execution, writing, and review.
xjtulyc/MedgeClaw 7 Biomedical research skills with real-time dashboard, RStudio, and JupyterLab integration.
zamushwani2/biomedical-ai-skills 4 Cancer multi-omics analysis skills in R.
ArcInstitute/SRAgent 1 Intelligent SRA and GEO dataset retrieval.
NVIDIA-BioNeMo/bionemo-agent-toolkit 17 Official NVIDIA BioNeMo NIM skills: Boltz-2, DiffDock, OpenFold2/3, RFdiffusion, ProteinMPNN, GenMol, Evo2, MolMIM, Parabricks.
JimLiu/science-skills 29 Reverse-engineered Claude Science built-in skills: structure prediction, design, genomics, single-cell, literature, and compute.
BioTender-max/awesome-bio-agent-skills 1 Self-referential hub skill that indexes this collection (browse & install entry point).

Skill Format

Each skill is a self-contained folder with a SKILL.md file defining domain knowledge, tool usage, and expected outputs. Compatible with any Claude-based agent framework supporting the SKILL.md convention.

skills/
└── /
    └── /
        ├── SKILL.md          # Skill definition (required)
        └── ...               # Supporting files

Quick install:

git clone https://github.com/BioTender-max/awesome-bio-agent-skills.git
cp -r awesome-bio-agent-skills/skills/* /path/to/your/agent/skills/

A machine-readable index of all 1,693 skills is available in bioskill_index_v3.csv.


Star History

If this collection is useful to you, please consider giving it a ⭐ — it helps others discover it.

View this README on GitHub

推荐工具

换一个关键词,或者移除筛选条件。

安装

npx skillfish add biotender-max/awesome-bio-agent-skills