CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

tooluniverse-protein-structure-retrieval

Protein structure retrieval from RCSB PDB, PDBe, AlphaFold, SWISS-MODEL, 3D-Beacons (cross-provider structure aggregator), and PDB-REDO (re-refined X-ray structures), with disambiguation, quality assessment (resolution, R-factor, pLDDT, coverage), and metadata. Distinguishes high-quality experimental (X-ray under 2 Angstrom) vs predicted vs homology-model vs medium-quality structures. Use for fetching protein structures, structure-quality comparison, batch model-availability checks across a gene list, checking every structure provider at once, and selecting structures for drug design or modeling.

Quality

88%

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

The canonical home for this skill is tooluniverse-protein-structure-retrieval in mims-harvard/ToolUniverse

SKILL.md
Quality
Evals
Security

Quality

Content

78%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 highly actionable and tool-dense with a clear phased workflow, but it is a monolithic single file: provider-specific detail that warrants separate reference files is inlined, and a few sections restate domain knowledge Claude already has. Splitting provider guides into references/ and trimming the duplicate Tool Reference would lift both conciseness and progressive disclosure.

Suggestions

Move the SWISS-MODEL, 3D-Beacons, and PDB-REDO sections (with their verified-live examples) and the Tool Reference into references/ files (e.g., references/providers.md, references/tool-reference.md), keeping SKILL.md as a concise overview with clearly signaled one-level-deep links.

Trim the 'Domain Reasoning' section and the quality-tier/pLDDT tables to only the skill-specific thresholds not already standard knowledge, or fold them into the report template they calibrate.

Make the Phase 2 → Phase 3 transition an explicit validate-then-report loop (e.g., re-query when a structure is obsoleted or when quality scores are missing) instead of relying on the implicit 'LOOK UP DON'T GUESS' directive.

DimensionReasoningScore

Conciseness

The bulk is dense, non-obvious, tool-specific knowledge (verified-live API behaviors, fallback chains, parameter semantics like provider="pdb" isolating re-indexed entries). However, the 'Domain Reasoning' section and quality-tier/pLDDT tables restate standard structural biology Claude already knows, and the Tool Reference section re-lists tools already shown in Phase 2 code.

4 / 5

Actionability

Fully executable copy-paste-ready tool calls with real function names, parameters, and expected return fields (e.g., PDBeValidation_get_quality_scores(pdb_id=pdb_id), SwissModel_get_models(uniprot_id=..., provider="swissmodel"), data.pdb_entries[]), plus verified-live examples covering edge cases like null QMEAN scores.

5 / 5

Workflow Clarity

Clear phase sequence (Clarify → Disambiguate → Retrieve → Report) with explicit clarification criteria, fallback chains, an error-handling table, and the 'LOOK UP DON'T GUESS' verification directive. Checkpoints are mostly present but feedback loops (validate → fix → retry) are implicit rather than explicit, keeping it below the anchor-5 example.

4 / 5

Progressive Disclosure

No bundle files exist and all content is inlined in a ~175-line single file. Section headers are clear and navigable, but provider deep-dives (SWISS-MODEL, 3D-Beacons, PDB-REDO with verified-live examples) and the Tool Reference are content that belongs in one-level-deep references/ files for a skill this length.

3 / 5

Total

16

/

20

Passed

Description

92%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 description: concrete and comprehensive on capabilities, with an explicit 'Use for' clause covering natural trigger scenarios. Only minor room to grow on synonym coverage of natural user phrasings.

DimensionReasoningScore

Specificity

The description lists multiple concrete actions — retrieval across six named providers, disambiguation, quality assessment with specific metrics (resolution, R-factor, pLDDT, coverage), metadata, and quality-tier distinction (experimental vs predicted vs homology-model). Coverage is comprehensive with no notable gaps, matching the anchor-5 example structure.

5 / 5

Completeness

It explicitly answers both what (retrieval, disambiguation, quality assessment, metadata, quality-tier distinction across named providers) and when ('Use for fetching protein structures, structure-quality comparison, batch model-availability checks... selecting structures for drug design or modeling') with concrete trigger phrases.

5 / 5

Trigger Term Quality

Natural trigger phrases are present ('fetching protein structures', 'structure-quality comparison', 'batch model-availability checks across a gene list', 'selecting structures for drug design or modeling'). A few natural user phrasings are missing (e.g., 'download a PDB file', '3D structure', 'cryo-EM'), so it falls just short of the comprehensive-with-synonyms anchor.

4 / 5

Distinctiveness Conflict Risk

It occupies a clear niche — multi-provider protein structure retrieval with quality assessment — and names its providers (RCSB PDB, PDBe, AlphaFold, SWISS-MODEL, 3D-Beacons, PDB-REDO), making conflicts with other skills unlikely.

5 / 5

Total

19

/

20

Passed

Validation

100%

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

mims-harvard/ToolUniverse

Path

plugins/tooluniverse/skills/tooluniverse-protein-structure-retrieval/SKILL.md

Revision

37bb81a

Reviewed

Table of Contents

DiscoveryImplementationValidation

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.

Claim skill