Does this skill support local model execution?

Yes, it supports local usage of models such as esm3-sm-open-v1 and esmc-300m via the ESM Python SDK and Flash Attention.

Can I use the Forge API with this skill?

Absolutely, the skill includes implementation patterns for connecting to the Forge API for high-throughput and large-scale model inference, including ESM3-Large.

How do I install the necessary dependencies?

You can install the core library using 'uv pip install esm'. For faster local inference, it is recommended to also install 'flash-attn'.

What is the ESM Claude Code Skill?

It is a specialized capability that allows Claude to assist with protein design, sequence generation, and structure prediction using Evolutionary Scale Modeling tools like ESM3 and ESM C.

What tasks are best suited for ESM C versus ESM3?

Use ESM3 for generative design, structure prediction, and functional conditioning. Use ESM C for efficient protein embeddings, representation learning, and similarity analysis.

Scientific Protein Modeling (ESM)

Name: Scientific Protein Modeling (ESM)
Author: davila7

bydavila7

•

14,610

Data Science & ML

Facilitates advanced protein design, structure prediction, and representation learning using the Evolutionary Scale Modeling (ESM) toolkit.

About

This skill integrates Evolutionary Scale Modeling (ESM) capabilities directly into your development workflow, providing a comprehensive toolkit for protein engineering and bioinformatics. It enables users to perform complex tasks such as generative multimodal protein design across sequence, structure, and function, as well as high-efficiency embedding generation using ESM C. Whether you are conducting local experimentation with open weights or utilizing the scalable Forge API for production-grade inference, this skill streamlines the process of designing novel proteins, predicting functional annotations, and extracting deep biological representations for downstream machine learning tasks.

Key Features

Advanced structure prediction and inverse folding for protein design
High-quality protein embeddings for downstream ML tasks with ESM C
Chain-of-thought generation for iterative refinement of complex protein structures
Seamless integration with both local models and the cloud-based Forge API
Multimodal protein generation across sequence, structure, and function using ESM3
14,610 GitHub stars

Use Cases

Extracting protein representations for functional prediction or similarity analysis
Performing large-scale protein variant screening and sequence optimization
Designing novel proteins with specific functional properties or structural characteristics

About

Key Features

Advanced structure prediction and inverse folding for protein design
High-quality protein embeddings for downstream ML tasks with ESM C
Chain-of-thought generation for iterative refinement of complex protein structures
Seamless integration with both local models and the cloud-based Forge API
Multimodal protein generation across sequence, structure, and function using ESM3
14,610 GitHub stars

Use Cases

Extracting protein representations for functional prediction or similarity analysis
Performing large-scale protein variant screening and sequence optimization
Designing novel proteins with specific functional properties or structural characteristics