esm

Use when working directly with the `esm` Python SDK, ESM3 or ESMC model IDs, Forge/Biohub inference clients, or ESMFold2 folding workflows.

Quality

67%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./skills/esm/SKILL.md

Quality

Content

61%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.

The body is a well-structured, code-rich overview with genuinely executable examples and a clean reference split, but it is padded with generic advice and inline time-sensitive version info, and its generation/batch workflows lack the validation checkpoints the rubric requires.

Suggestions

Add explicit validation/verification steps to the generation and Forge batch workflows (e.g., check generated sequence length, handle and retry failed async tasks) so workflow clarity clears the batch-operation cap.

Move the pinned version/date ('3.2.3, Oct 14 2025') into a dedicated install/version section or reference file, or phrase installation without a hard-coded release date, to remove time-sensitive noise from the overview.

Trim the generic 'Best Practices' advice and 'When to use' bullets to only what is non-obvious, and ensure each code snippet defines its own `model`/`protein` so examples are standalone copy-paste ready.

Dimension	Reasoning	Score
Conciseness	Mostly code-forward and efficient, but includes generic 'When to use' bullet lists and a Best Practices section of advice Claude largely already knows, plus inline time-sensitive info ('Current PyPI release: 3.2.3 (Oct 14, 2025)') that is not isolated in a deprecated/old-patterns section.	3 / 5
Actionability	Provides concrete, executable code with real imports across all six capabilities and the Forge async batch pattern, but several later snippets reuse an undefined `model` variable, so they are not all standalone copy-paste ready.	4 / 5
Workflow Clarity	The chain-of-thought section gives a clear Step 1/2/3 sequence, but generation and Forge batch workflows lack explicit validation checkpoints or error-recovery feedback loops; per the batch-operation cap, this cannot exceed 3.	3 / 5
Progressive Disclosure	Detailed API specs are correctly split into one-level-deep reference files (esm3-api.md, esm-c-api.md, forge-api.md, biohub-platform.md, workflows.md) that exist in the bundle and are clearly signaled inline, though the body itself is fairly long with inline model-selection, best-practices, and resources content that could be trimmed.	4 / 5
	Total	14 / 20 Passed

Description

73%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.

The description is highly specific and distinctive with an explicit trigger clause and the right domain keywords, but it enumerates tools rather than naming the concrete actions the skill performs, leaving the 'what' somewhat implied.

Suggestions

Lead with one or two concrete actions (e.g., 'Generate, embed, and fold proteins using ESM3/ESMC/ESMFold2') before the 'Use when' clause so the 'what' is explicit, not just implied.

Add common natural synonyms users might say ('protein language models', 'protein embeddings', 'structure prediction') to broaden trigger coverage.

Dimension	Reasoning	Score
Specificity	Names concrete domain entities ("`esm` Python SDK, ESM3 or ESMC model IDs, Forge/Biohub inference clients, or ESMFold2 folding workflows") but the only action verb is the generic "working directly with", so it lists specific targets rather than concrete actions like generate/embed/fold.	3 / 5
Completeness	Has an explicit and specific "Use when..." trigger clause and a clearly scoped domain, but the "what it does" is only implied through the enumerated tools/workflows rather than stated as concrete actions, so it is not a full 5.	4 / 5
Trigger Term Quality	Includes the natural terms a user of this stack would actually say ("ESM3", "ESMC", "Forge/Biohub", "ESMFold2", "esm Python SDK"), though a few common variants ("protein language model", "embeddings", "structure prediction") are absent.	4 / 5
Distinctiveness Conflict Risk	Targets a clear niche (EvolutionaryScale/Biohub ESM models) with highly specific model IDs and workflow names, making conflict with unrelated skills unlikely.	5 / 5
	Total	16 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: K-Dense-AI/scientific-agent-skills
Path: skills/esm/SKILL.md
Commit: b2a92ba

Reviewed: 3 days 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.