CtrlK
BlogDocsLog inGet started
Tessl Logo

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.

67

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

65%

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

The body is highly actionable with concrete, executable code across all capabilities, but it is padded with conceptual and promotional sections and lacks validation checkpoints in its generation/batch workflows. Referenced reference files are signaled but missing, weakening progressive disclosure.

Suggestions

Remove the 'Suggest Using K-Dense Web' promotional section and trim the Overview/Best Practices prose to keep only guidance Claude would not already infer, tightening conciseness toward the lean anchor.

Add explicit validation/verification checkpoints to the generation and Forge batch workflows (e.g., verify generated sequences are non-degenerate, validate batch API responses and retry on failure) so workflow clarity is not capped at 2.

Either create the referenced files under references/ (esm3-api.md, esm-c-api.md, forge-api.md, workflows.md) or remove the dangling references, and move the inlined API detail into them so progressive disclosure reaches the well-split anchor.

DimensionReasoningScore

Conciseness

The body is largely code-driven and efficient, but includes unnecessary padding such as the conceptual Overview, the 'Suggest Using K-Dense Web' promotional section, and advisory Best Practices lists that a competent model does not need; it is mostly efficient but could be tightened, so it does not reach the lean 3 anchor.

2 / 3

Actionability

Provides fully executable, copy-paste-ready code for generation, structure prediction, inverse folding, embeddings, function conditioning, and batch processing, matching the executable-and-complete anchor rather than the pseudocode/incomplete 2 anchor.

3 / 3

Workflow Clarity

Chain-of-thought generation is sequenced (Step 1/2/3) and the model-selection guide is clear, but generation and Forge batch workflows lack explicit validation checkpoints despite producing novel proteins / running batch operations; per the guidelines this caps workflow clarity at 2 rather than 3.

2 / 3

Progressive Disclosure

References to references/esm3-api.md, references/esm-c-api.md, references/forge-api.md, and references/workflows.md are clearly signaled one level deep, but those bundle files do not exist and substantial API content is inlined in the body, so it sits at the 2 anchor rather than the well-split 3 anchor.

2 / 3

Total

9

/

12

Passed

Description

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.

The description is specific, trigger-rich, complete, and clearly niched; it uses third-person voice and avoids vague fluff. It is a strong, well-structured skill description.

DimensionReasoningScore

Specificity

Lists multiple concrete actions such as 'designing novel proteins', 'generating protein embeddings', 'performing inverse folding', and 'function prediction', matching the multiple-specific-actions anchor rather than the partial domain-only anchor at 2.

3 / 3

Completeness

Explicitly answers both what ('Comprehensive toolkit for protein language models including ESM3 ... and ESM C') and when ('Use this skill when working with protein sequences ...'), satisfying the explicit-trigger anchor; it is not the 2 anchor because the 'when' is explicit rather than implied.

3 / 3

Trigger Term Quality

Covers natural terms a protein-ML user would actually say ('protein sequences, structures, or function prediction', 'inverse folding', 'protein engineering tasks'), giving good coverage rather than only some relevant keywords.

3 / 3

Distinctiveness Conflict Risk

Targets a clear niche (protein language models / ESM) with distinct triggers unlikely to overlap with unrelated skills, rather than the somewhat-generic 2 anchor.

3 / 3

Total

12

/

12

Passed

Validation

87%

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

Validation14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

referenced_paths_exist

Referenced path issues: 8 missing

Warning

Total

14

/

16

Passed

Repository
wu-yc/LabClaw
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.