alphafold-database-fetch-and-analyze

Retrieve and analyze AlphaFold predicted structures for a protein. Use when the user provides a specific UniProt Accession ID and wants structural confidence metrics (pLDDT), domain boundary analysis, or disorder assessment. Do not use if the user only has a protein name, gene name, or amino acid sequence — ask for a UniProt ID first.

Quality

96%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

92%Weight 40%Scale 1-3

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

A lean, actionable skill body with concrete executable examples and clear validation checkpoints. The main gap is progressive disclosure: the references/citation.bib bundle file is provided but never signaled from the skill, leaving it orphaned.

Suggestions

Add a one-line pointer to references/citation.bib (e.g., in Prerequisites or a References section) so the bundled citation file is discoverable and Claude knows when to use it.

State the intended run order of the three scripts explicitly (fetch → analyze_plddt + analyze_pae → interpret) to make the data-flow pipeline unambiguous rather than implied.

Consider trimming the duplicated "Do NOT use when" list in Overview since the description frontmatter already carries a "Do not use if" guard, keeping only the body-specific distinctions (Foldseek, RCSB PDB, custom-sequence AlphaFold).

Dimension	Reasoning	Score
Conciseness	The body is lean: it never explains what AlphaFold, pLDDT, or mmCIF are, and assumes Claude's competence. Sections earn their place; it is not the verbose concept-padded anchor, and the slight repetition of the anti-trigger list is the body expanding the description rather than fluff.	3 / 3
Actionability	Provides copy-paste-ready executable commands with real identifiers — e.g. "uv run scripts/fetch_structure.py P00520 -o /path/to/output/" and "uv run scripts/analyze_plddt.py ./data/AF-P00520-F1-metadata.json" — plus concrete path guidance, matching the fully-executable anchor.	3 / 3
Workflow Clarity	A clear fetch → analyze → interpret sequence with explicit validation checkpoints: the license-file existence check in Prerequisites and the mandatory "Check the script output for any [!] WARNING lines" step in Interpreting the Output form a checklist for a non-destructive fetch/analyze task.	3 / 3
Progressive Disclosure	The three scripts are well-signaled one level deep, but references/citation.bib exists as a bundle file and is never referenced or navigated to from the body — a reference present but not clearly signaled, which the bundle-structure guideline penalizes.	2 / 3
	Total	11 / 12 Passed

Description

100%Weight 40%Scale 1-3

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 description: concrete capabilities, an explicit "Use when" trigger tied to a UniProt ID, and a clear "Do not use if" anti-trigger that distinguishes it from neighboring skills. No vague fluff or over-claims.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "Retrieve and analyze AlphaFold predicted structures", "structural confidence metrics (pLDDT), domain boundary analysis, or disorder assessment" — matching the anchor for listing several specific concrete actions rather than vague language.	3 / 3
Completeness	Explicitly answers both what (retrieve/analyze structures, pLDDT, domains, disorder) and when ("Use when the user provides a specific UniProt Accession ID..."), plus a "Do not use if..." anti-trigger, matching the top anchor.	3 / 3
Trigger Term Quality	Includes natural domain terms a user would actually say — "AlphaFold", "UniProt Accession ID", "pLDDT", "domain boundary", "disorder" — giving good coverage rather than only technical jargon.	3 / 3
Distinctiveness Conflict Risk	A clear niche (AlphaFold DB structures keyed by UniProt ID) with an explicit "Do not use if the user only has a protein name, gene name, or amino acid sequence" guard, making it unlikely to trigger for the wrong skill.	3 / 3
	Total	12 / 12 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: google-deepmind/science-skills
Path: skills/alphafold_database_fetch_and_analyze/SKILL.md
Commit: 0b42509

Reviewed: 8 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.