bioSkills

A collection of skills that guide AI coding agents (Claude Code, Codex, Gemini) through common bioinformatics tasks.

Project Goal

This repository provides AI agents with expert knowledge for bioinformatics workflows. Each skill contains code patterns, best practices, and examples that help agents generate correct, idiomatic code for common tasks.

Target users range from undergrads learning computational biology to PhD researchers processing large-scale data. The skills cover the full spectrum from basic sequence manipulation to advanced analyses like single-cell RNA-seq and population genetics.

Requirements

Python

Python 3.9+
biopython, pysam, cyvcf2, pybedtools, pyBigWig, scikit-allel, anndata

pip install biopython pysam cyvcf2 pybedtools pyBigWig scikit-allel anndata mygene

R/Bioconductor

Required for differential expression, single-cell, pathway analysis, and methylation skills.

if (!require('BiocManager', quietly = TRUE))
    install.packages('BiocManager')
BiocManager::install(c('DESeq2', 'edgeR', 'Seurat', 'clusterProfiler', 'methylKit'))

CLI Tools

# macOS
brew install samtools bcftools blast minimap2 bedtools

# Ubuntu/Debian
sudo apt install samtools bcftools ncbi-blast+ minimap2 bedtools

# conda
conda install -c bioconda samtools bcftools blast minimap2 bedtools \
    fastp kraken2 metaphlan sra-tools bwa-mem2 bowtie2 star hisat2 \
    manta delly cnvkit macs3 tobias

Installation

Claude Code

git clone https://github.com/your-username/bioSkills.git
cd bioSkills
./install-claude.sh                              # Install globally
./install-claude.sh --project /path/to/project   # Or install to specific project
./install-claude.sh --list                       # List available skills
./install-claude.sh --validate                   # Validate all skills
./install-claude.sh --update                     # Only update changed skills
./install-claude.sh --uninstall                  # Remove all bio-* skills

Codex CLI

./install-codex.sh                               # Install globally
./install-codex.sh --project /path/to/project    # Or install to specific project
./install-codex.sh --validate                    # Validate all skills
./install-codex.sh --update                      # Only update changed skills
./install-codex.sh --uninstall                   # Remove all bio-* skills

Gemini CLI

./install-gemini.sh                              # Install globally
./install-gemini.sh --project /path/to/project   # Or install to specific project
./install-gemini.sh --validate                   # Validate all skills
./install-gemini.sh --update                     # Only update changed skills
./install-gemini.sh --uninstall                  # Remove all bio-* skills

Codex and Gemini installers convert to the Agent Skills standard (examples/ -> scripts/, usage-guide.md -> references/).

Skill Categories

Category	Skills	Primary Tools	Description
sequence-io	9	Bio.SeqIO	Read, write, convert FASTA/FASTQ/GenBank and 40+ formats
sequence-manipulation	7	Bio.Seq, Bio.SeqUtils	Transcription, translation, motif search, sequence properties
database-access	10	Bio.Entrez, BLAST+, SRA toolkit, UniProt API	NCBI/UniProt queries, SRA downloads, BLAST, homology searches
alignment-files	9	samtools, pysam	SAM/BAM/CRAM viewing, sorting, filtering, statistics, validation
variant-calling	13	bcftools, cyvcf2, Manta, Delly, VEP, SnpEff	VCF/BCF calling, SVs, filtering, annotation, clinical interpretation
alignment	4	Bio.Align, Bio.AlignIO	Pairwise and multiple sequence alignment, MSA statistics, alignment I/O
phylogenetics	5	Bio.Phylo, IQ-TREE2, RAxML-ng	Tree I/O, visualization, ML inference with model selection, ultrafast bootstrap
differential-expression	6	DESeq2, edgeR, ggplot2, pheatmap	RNA-seq differential expression, visualization, batch correction
structural-biology	6	Bio.PDB, ESMFold, Chai-1	PDB/mmCIF parsing, SMCRA navigation, geometric analysis, ML structure prediction
single-cell	13	Seurat, Scanpy, Pertpy, Cassiopeia	scRNA-seq QC, clustering, trajectory, communication, annotation, perturb-seq, lineage tracing
pathway-analysis	6	clusterProfiler, ReactomePA, rWikiPathways, enrichplot	GO, KEGG, Reactome, WikiPathways enrichment
restriction-analysis	4	Bio.Restriction	Restriction sites, mapping, enzyme selection
methylation-analysis	4	Bismark, methylKit, bsseq	Bisulfite alignment, methylation calling, DMRs
chip-seq	7	MACS3, ChIPseeker, DiffBind	Peak calling, annotation, differential binding, motifs, QC, super-enhancers
metagenomics	7	Kraken2, MetaPhlAn, Bracken, HUMAnN	Taxonomic classification, abundance estimation, func

...

gptomics/bioskills

README