esm

Toolkit for protein language models (ESM3 for multimodal generative protein design; ESM C for efficient embeddings). Use when you need sequence/structure/function generation or prediction, inverse folding, protein embeddings, or scalable inference via local weights or the Forge API.

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

65%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

Well-structured and actionable with good progressive disclosure and real reference files, but held back by overlapping overview sections, some basic-concept padding, and a batch Forge workflow lacking validation/verification steps.

Suggestions

Collapse the 'When to Use' and 'Key Features' sections into one to remove the restated-capability redundancy, and drop explanations of concepts Claude already knows (e.g., the temperature diversity note).

Add a validation/verification step to the Forge batch workflow — check each result for errors, handle failures with a retry, and confirm outputs before reporting success — so the batch operation has a feedback loop.

Make the ESM C embedding example concrete: show the actual embedding tensor shape/access (e.g., embeddings.shape or slicing) instead of only printing 'embeddings computed.'

Dimension	Reasoning	Score
Conciseness	Mostly efficient but could be tightened: the 'When to Use' and 'Key Features' sections restate the same capabilities, and basic-concept explanations (e.g., 'temperature: controls sampling diversity (lower = more deterministic; higher = more diverse)') pad text Claude already knows.	3 / 5
Actionability	Provides concrete, mostly executable code with specific model names and API calls across local ESM3, Forge batch, and ESM C; the ESM C example only prints 'embeddings computed' and defers the actual tensor shape, a minor gap.	4 / 5
Workflow Clarity	The Forge example is a batch operation (async_generate of 5 proteins) with no validation, result-checking, or error-recovery feedback loop, which per the rubric caps workflow clarity at 3 even though a rough sequence is present.	3 / 5
Progressive Disclosure	Clear overview in SKILL.md with well-signaled, one-level-deep references (esm3-api.md, esm-c-api.md, forge-api.md, workflows.md — all verified present), with detailed API material appropriately split out rather than inlined.	5 / 5
	Total	15 / 20 Passed

Description

100%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, specific description that clearly states capabilities and provides an explicit 'Use when' trigger clause with natural domain terms. Both the 'what' and 'when' are answered concretely with minimal conflict risk.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — sequence/structure/function generation or prediction, inverse folding, protein embeddings, and scalable inference via local weights or Forge — giving comprehensive coverage of the toolkit's capabilities.	5 / 5
Completeness	Explicitly answers both what ('Toolkit for protein language models (ESM3...; ESM C...)') and when ('Use when you need sequence/structure/function generation or prediction, inverse folding, protein embeddings, or scalable inference...').	5 / 5
Trigger Term Quality	Natural domain phrases a user would say ('protein embeddings', 'inverse folding', 'structure prediction', 'generative protein design') are covered with synonyms ('generation or prediction', 'local weights or the Forge API').	5 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (ESM3/ESM C protein language models, Forge API) with distinct triggers and minimal overlap risk against other skills.	5 / 5
	Total	20 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: aipoch/medical-research-skills
Path: scientific-skills/Protocol Design/esm/SKILL.md
Commit: 63c61d3

Reviewed: 5 hours ago

Table of Contents

Discovery Implementation Validation

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.