esm

Design and analyze proteins with ESM3/ESM C models via local or Forge inference.

94|11|Updated Mar 26, 2026
One-click install
npx skills add https://github.com/swaruplab/operon --skill esm-swaruplab
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: esm
Source: https://github.com/swaruplab/operon/tree/main/src-tauri/protocols/esm
Command: npx skills add https://github.com/swaruplab/operon --skill esm-swaruplab

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

ESM provides state-of-the-art protein language models for understanding, generating, and designing proteins, enabling researchers to design novel proteins and analyze embeddings without deep ML expertise.

Core Features & Use Cases

  • ESM3 multimodal generation across sequence, structure, and function for design, refinement, and exploration.
  • Protein embeddings and representations with ESM C for fast, scalable feature extraction and downstream analysis.
  • End-to-end workflows including sequence completion, structure prediction, function conditioning, and batch processing via Forge for large-scale projects.

Quick Start

Provide a partial or masked protein sequence and run an initial generation using ESM3 with a suitable GenerationConfig to complete the design.

Frequently Asked Questions about esm

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I design a protein sequence using ESM3?

To design a protein with ESM3, provide a partial or masked sequence and execute an initial generation using a suitable GenerationConfig to complete the design. This enables multimodal generation across sequence, structure, and function.

Can I extract protein embeddings for downstream analysis without deep ML expertise?

Yes, you can generate protein embeddings for downstream analysis without deep ML expertise by using ESM C. It provides fast, scalable feature extraction and representations for end-to-end workflows.

Do I need local model weights to run ESM3 and ESM C inference?

You do not strictly need local model weights to run ESM3 and ESM C inference, as Forge-based access is supported. Optional model weights for esm3-sm-open-v1 or esm3-medium/large can be used for local inference.

Does this approach support batch processing for large-scale protein libraries?

Yes, this approach supports batch processing for large-scale protein libraries via Forge. It enables end-to-end workflows for sequence completion, structure prediction, and function conditioning across single proteins or libraries.

What is the best way to condition protein generation on specific functions?

The best way to condition protein generation on specific functions is by using ESM3 multimodal capabilities. It enables function conditioning alongside sequence design and structure prediction to refine and explore novel proteins.