TianGzlabOfficial·56 Agent Skills Included

OmicsClaw

Local-first multi-omics analysis across eight research domains

Runs 95 reproducible bioinformatics skills covering spatial transcriptomics, single-cell, genomics, proteomics, metabolomics, and bulk RNA-seq analysis. Eliminates scattered Python, R, and CLI pipelines by routing plain-language questions to the right analysis workflow automatically. Keeps raw data on your machine while producing publication-ready figures, tables, and reports with full session memory.

155|26

↗

npx skills add TianGzlab/OmicsClaw --all -g -y

Available:

Instructs the agent to answer omics questions only by routing them to the correct domain skill via a keyword routing table, and to ground every answer in a SKILL.md methodology or script output.

All Skills in This Repository (56)

Pure Emerald Level Indicators

📦 In Repo

TianGzlab

Literature Parsing Skill

Parse scientific literature to extract GEO dataset metadata and download associated data.

orchestrator

Route natural language queries to omics analysis skills across multiple domains.

metabolomics-annotation

Annotate metabolomics features by matching MS2 spectra against SIRIUS, GNPS, and MetFrag databases.

metabolomics-normalization

Normalize metabolomics data using median, quantile, total-ion-count, PQN, and log2 transformations.

metabolomics-de

Perform univariate and multivariate statistical analysis on metabolomics data to identify biomarkers.

metabolomics-quantification

Impute missing values and normalize metabolomics feature data.

metabolomics-statistics

Perform univariate statistical tests and FDR correction on metabolomics data.

metabolomics-xcms-preprocessing

Preprocess LC-MS/GC-MS metabolomics data with XCMS3 peak detection and alignment.

metabolomics-peak-detection

Detect metabolite peaks from LC-MS data using XCMS, MZmine 3 or MS-DIAL and extract features for downstream analysis.

metabolomics-pathway-enrichment

Perform metabolomics pathway enrichment with hypergeometric tests and FDR correction.

sc-grn

Infers gene regulatory networks from single-cell expression data using pySCENIC or correlation-based fallback.

sc-batch-integration

Integrate multi-sample single-cell RNA sequencing data with Harmony, scVI, Seurat CCA/RPCA, BBKNN, and fastMNN.

Official

Advanced

Frequently Asked Questions

FAQPage Schema

How to install OmicsClaw?▼

Run `npx skills add TianGzlab/OmicsClaw --all -g -y` in your terminal to install all skills in this suite globally.

What omics data types does OmicsClaw support?▼

It covers eight domains: spatial transcriptomics, single-cell omics, genomics, proteomics, metabolomics, bulk RNA-seq, orchestration, and literature parsing, with 95 registered analysis skills.

Does OmicsClaw keep my data private?▼

Yes. All processing is local-first, so raw matrices never leave your machine; remote Linux execution over SSH is optional for large server-side data.

Can I run analyses without writing code?▼

Yes. You describe the analysis in plain language and the orchestrator routes it to the right skill, which runs reproducible scripts and returns figures, tables, and reports.

Does OmicsClaw work with Claude Code and other agents?▼

Yes. Every skill follows the standard SKILL.md format and runs through a unified CLI (`oc run <skill>`), so it works in Claude Code, Cursor, and other compatible agent environments.