PharMolixCommunity·41 Agent Skills Included

OpenBioMed

AI skills for drug discovery, protein design, and single-cell analysis

Automates end-to-end biomedical research tasks including drug candidate generation, protein structure prediction, mutation analysis, and single-cell omics processing. Eliminates manual database lookups across PubChem, ChEMBL, KEGG, UniProt, and STRING by turning them into direct AI-driven queries. Provides 45 ready-made skills with pre-configured models, validation steps, and report templates so researchers complete complex workflows faster.

1,105|132

↗

npx skills add PharMolix/OpenBioMed --all -g -y

Available:

Tells the AI agent how the OpenBioMed toolkit is structured, which tasks, models, and tools are registered, and how to run training, inference, and workflow pipelines when executing the biomedical skills.

All Skills in This Repository (41)

Pure Emerald Level Indicators

📦 In Repo

PharMolix

antibody-structure-prediction-tfold

Predict antibody and nanobody structures from sequences using tfold-based models.

retrosynthesis-planning

Plan retrosynthetic routes for complex molecules using AND/OR tree search and vendor validation.

disease-drug-intelligence

Consolidate multi-database evidence to identify innovative drug opportunities for a specified disease.

molecule-biochemical-significance-query-biot5

Query biochemical significance of molecules via BioT5 from SMILES or names.

binding-affinity-prediction-prodigy

Predict protein complex binding affinity scores from PDB structures using Prodigy Prot.

structure-prediction-boltz-2

Predict protein and protein-ligand complex structures and affinities using Boltz-2.

text-based-molecule-editing

Edit molecular structures from natural language prompts using MolT5/BioT5 pipelines.

iupac-name-identification-biot5

Identify IUPAC names from SMILES strings or common names using BioT5, MolT5, and RDKit.

mutation-design-gfp

Design GFP mutant libraries through iterative optimization and export 96-sequence CSV files.

single-cell-foundation-model-langcell

Annotates cell identity by aligning embeddings with text descriptions and ontology context.

target-drug-report

Generate structured drug development reports for specified therapeutic targets.

antibody-design-iggm

Design antibodies with IgGM models using antigen PDB data.

Community

Advanced

Frequently Asked Questions

FAQPage Schema

How to install OpenBioMed?▼

Run `npx skills add PharMolix/OpenBioMed --all -g -y` in your terminal to install all 45 biomedical skills globally.

What can OpenBioMed do for drug discovery?▼

It generates drug candidates for a target, predicts ADMET and toxicity properties, runs docking and binding analysis, and plans retrosynthesis routes automatically.

Can OpenBioMed analyze proteins and antibodies?▼

Yes. It predicts protein and antibody structures, explains mutation effects, designs functional proteins and antibodies, and estimates binding affinity.

Does OpenBioMed support single-cell data analysis?▼

Yes. It includes skills for scRNA-seq clustering, cell type annotation with LangCell and scGPT, spatial transcriptomics, ATAC-seq, and proteomics pipelines.

Do I need coding experience to use OpenBioMed skills?▼

No. Once installed, your AI agent runs the underlying models and database queries based on your plain-English research questions.