esm

Comprehensive toolkit for protein language models including ESM3 (generative multimodal protein design across sequence, structure, and function) and ESM C (efficient protein embeddings and representations). Use this skill when working with protein sequences, structures, or function prediction; designing novel proteins; generating protein embeddings; performing inverse folding; or conducting protein engineering tasks. Supports both local model usage and cloud-based Forge API for scalable inference.

3.44x

Quality

86%

Does it follow best practices?

Impact

93%

3.44x

Average score across 3 eval scenarios

Securityby

Passed

No known issues

Evaluation results

79%

45%

Protein Functional Similarity Analysis

ESM C embedding pipeline

Criteria

Without context

With context

ESMC import

100%

ESMProtein import

100%

from_pretrained loading

Correct model selection

100%

Device placement

50%

encode step

100%

forward step

20%

Mean pooling

83%

100%

Cosine similarity

100%

Output JSON report

100%

Embeddings saved

100%

91%

De Novo Enzyme Design with Multimodal Protein Generation

ESM3 chain-of-thought protein design

Criteria

Without context

With context

ESM3 import

100%

API imports

100%

Local model ID

100%

from_pretrained loading

100%

Underscore masking

100%

Sequence track

100%

Structure track

100%

Multi-track iteration

100%

num_steps scaling

100%

Temperature parameter

100%

PDB output

37%

100%

FASTA output

100%

60%

High-Throughput Protein Variant Screening Pipeline

Forge API batch processing with error handling

Criteria

Without context

With context

Forge client import

100%

SDK API imports

100%

Client initialization

100%

Production model

100%

Token from env var

100%

Async generation

100%

asyncio.gather concurrency

100%

Rate limit retry

100%

Exponential backoff

100%

Underscore masking preserved

100%

FASTA output

100%

Repository: K-Dense-AI/claude-scientific-skills
Commit: b58ad7e

Evaluated: 27 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Protein Functional Similarity Analysis De Novo Enzyme Design with Multimodal Protein Generation High-Throughput Protein Variant Screening Pipeline

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.