Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
This collection provides structured guidance for automating the preprocessing of biological sequence data. It assists bioinformatics engineers in managing complex alignment tasks, from sorting and indexing BAM files to cleaning amplicon-panel data for variant calling.
By focusing on precise tool implementation, the repository helps researchers remove primer-derived bases and filter alignments to ensure that reference footprints do not introduce false positives in clinical or research assays.
WHO IT'S FOR
Bioinformatics engineers
automating alignment file preprocessing with AI agents
Genomics researchers
cleaning amplicon-panel data for variant calling
Computational biologists
validating alignment quality before downstream analysis
AI automation builders
configuring AI agents for bioinformatics workflows
Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.
Installs: Full bioSkills collection installed globally to ~/.gemini/antigravity/skills/; the clawhub-installer meta-skill is excluded. · Antigravity CLI
Before you start
Git with SSH access to git@github.com:GPTomics/bioSkills.git
A Bash-compatible environment
Run from: bioSkills repository root
In a terminal, clone the repository.
git clone git@github.com:GPTomics/bioSkills.git
In the terminal, enter the cloned repository.
cd bioSkills
From the repository root, run the Antigravity CLI installer in its default global mode.
Path to an input BAM file, such as [path/to/input.bam]
Desired minimum mapping quality
Any reads to include or exclude by FLAG
Optional BED file or genomic regions
samtools 1.19+ for the documented reference patterns
Use the bio-alignment-filtering skill to propose a samtools view command for [path/to/input.bam] that keeps the requested high-quality alignments. Explain the selected -F, -f, -q, or -L filters briefly, but do not run the command. Ask me for any missing MAPQ threshold, FLAG requirements, or BED region path.
Path to a table containing p-values, such as [path/to/p_values.csv]
Name of the p-value column
Whether the analysis is discovery or confirmatory
Known dependence assumptions, if any
R or Python with a documented compatible implementation, such as R p.adjust or statsmodels 0.14+
Use the bio-experimental-design-multiple-testing skill to review the p-values in [path/to/p_values.csv]. First determine whether this is discovery or confirmatory analysis and whether arbitrary dependence is a concern, then recommend one supported correction method and produce adjusted values with a short interpretation of the controlled error rate.
A preprocessed single-cell object, such as [path/to/preprocessed.h5ad], or an equivalent Seurat object
Whether to use Scanpy or Seurat
Scanpy 1.10+ with anndata 0.10+, or Seurat 5.0+
Any known batch or biological covariates
Use the bio-single-cell-clustering skill on [path/to/preprocessed.h5ad] to create an initial k-nearest-neighbor graph, Leiden clustering, and UMAP. State the chosen number of PCs, neighbors, resolution, backend, iteration count, and random seed, and flag the clustering as a hypothesis requiring marker and stability validation.
Claude
Codex
Cursor
Gemini CLI
OpenClaw
Amp
OpenCode
Antigravity
Cline