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.
89
86%
Does it follow best practices?
Impact
93%
3.44xAverage score across 3 eval scenarios
Passed
No known issues
Quality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is an excellent skill description that clearly identifies the domain (protein language models), names specific tools (ESM3, ESM C), lists concrete capabilities, and provides explicit trigger guidance via a 'Use this skill when...' clause. It uses proper third-person voice throughout and covers both what the skill does and when to use it comprehensively. The highly specialized domain makes it very unlikely to conflict with other skills.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'generative multimodal protein design across sequence, structure, and function', 'efficient protein embeddings and representations', 'inverse folding', 'protein engineering tasks', 'function prediction'. Names specific models (ESM3, ESM C) and deployment modes (local, Forge API). | 3 / 3 |
Completeness | Clearly answers both 'what' (comprehensive toolkit for protein language models with specific capabilities listed) and 'when' (explicit 'Use this skill when...' clause covering multiple trigger scenarios like working with protein sequences, designing novel proteins, generating embeddings, inverse folding, and protein engineering). | 3 / 3 |
Trigger Term Quality | Includes strong natural keywords users would say: 'protein sequences', 'protein structures', 'function prediction', 'designing novel proteins', 'protein embeddings', 'inverse folding', 'protein engineering', 'ESM3', 'ESM C', 'protein language models'. These cover the natural vocabulary of the domain well. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive niche — protein language models, ESM3/ESM C, protein design, inverse folding, and protein engineering are very specific domain terms unlikely to conflict with other skills. The named models and domain-specific terminology create a clear, unique trigger profile. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
72%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured skill with strong actionability through executable code examples covering all major ESM capabilities, and excellent progressive disclosure with clear references to detailed documentation. The main weaknesses are moderate verbosity (explanatory 'When to use' sections, best practices that could be trimmed) and lack of validation/error-handling checkpoints in the workflows, which is notable for generative protein design tasks where outputs should be verified.
Suggestions
Add explicit validation checkpoints to workflows (e.g., check if generation returned valid sequences, verify coordinate outputs are not None, validate sequence length matches expectations) to improve workflow clarity.
Trim or remove the 'When to use' bullet lists under each capability—Claude can infer appropriate use cases from the code examples and section titles.
Remove the 'Responsible Use' and 'Resources and Documentation' sections or move them to a reference file—these consume tokens without providing actionable guidance for task execution.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill includes some unnecessary explanatory text (e.g., 'When to use' lists that Claude can infer, the 'Responsible Use' section, and verbose descriptions of model capabilities). However, the code examples are reasonably tight and the model selection guide is useful reference material that earns its place. | 2 / 3 |
Actionability | The skill provides fully executable Python code examples for all major use cases: sequence generation, structure prediction, inverse folding, embeddings, function conditioning, chain-of-thought generation, and batch processing. Code is copy-paste ready with proper imports and realistic usage patterns. | 3 / 3 |
Workflow Clarity | The chain-of-thought generation section shows a clear multi-step workflow, but there are no validation checkpoints or error recovery steps anywhere. For protein design tasks involving generation and structure prediction, there should be explicit validation steps (e.g., checking generation success, validating output coordinates, verifying sequence plausibility) but these are absent. | 2 / 3 |
Progressive Disclosure | Excellent progressive disclosure with a clear overview in the main file and well-signaled one-level-deep references to esm3-api.md, esm-c-api.md, forge-api.md, and workflows.md. The main file provides enough to get started while pointing to detailed references for advanced usage. | 3 / 3 |
Total | 10 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
metadata_version | 'metadata.version' is missing | Warning |
Total | 10 / 11 Passed | |
- Repository
- K-Dense-AI/claude-scientific-skills
- Commit
25e1c0f
- Reviewed
Table of Contents
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.