alphafold-database

Access AlphaFold 200M+ AI-predicted protein structures. Retrieve structures by UniProt ID, download PDB/mmCIF files, analyze confidence metrics (pLDDT, PAE), for drug discovery and structural biology.

Quality

65%

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

Fix and improve this skill with Tessl

tessl review fix ./backend/cli/skills/databases/alphafold-database/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 strong on actionability, with concrete, executable examples across all access methods, and a real, well-signaled reference file. It is dragged down by noticeable redundancy (confidence-metric guidance repeated three times), padding sections, and the absence of validation checkpoints in batch and bulk-download workflows.

Suggestions

Consolidate the pLDDT/PAE interpretation guidance (currently in section 3 comments, "Key Concepts", and "Confidence Interpretation Guidelines") into a single section and cut the generic "Common Use Cases" bullets.

Add validation checkpoints to batch and bulk workflows, e.g. check response.status_code before writing files and verify download completeness after gsutil copies.

Move detailed BigQuery/gsutil/3D-Beacons material into references/api_reference.md and convert the Resources pointer into a markdown link.

Dimension	Reasoning	Score
Conciseness	Mostly efficient, code-driven content, but pLDDT thresholds appear three times (section 3 code comments, "Key Concepts", "Confidence Interpretation Guidelines"), PAE guidance and v4 version notes each appear twice, and the "Common Use Cases" section is generic bullet padding — the body could be tightened considerably.	3 / 5
Actionability	Largely executable, copy-paste-ready code covering Biopython retrieval, REST endpoints, file downloads, gsutil, BigQuery, and structure parsing; minor gaps include an undefined alphafold_id in the section 3 PAE block and missing imports (requests, np) in the batch-processing example.	4 / 5
Workflow Clarity	Content is organized by capability rather than a sequenced workflow, and batch/bulk operations (proteome downloads, batch processing) lack validation checkpoints such as HTTP status checks or download verification — the batch-operation rule caps this at 3 despite the try/except in the batch example.	3 / 5
Progressive Disclosure	The references/api_reference.md file exists, is one level deep, and is clearly signaled with a content summary and consult-conditions; however, ~500 lines of inlined detail (BigQuery, gsutil, 3D-Beacons) that could live in the reference, and a plain-text rather than markdown link to it, leave minor organization gaps.	4 / 5
	Total	14 / 20 Passed

Description

70%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 specific and distinctive, naming concrete actions and domain-specific trigger terms (UniProt, pLDDT, PAE). Its main weakness is the absence of an explicit "Use when..." clause, leaving the "when to use" guidance only weakly implied by application domains.

Suggestions

Append an explicit trigger clause, e.g. "Use when the user mentions AlphaFold, predicted protein structures, pLDDT/PAE confidence, or needs structure files for a UniProt accession."

Mention bulk/proteome-scale access (e.g., "download whole-proteome datasets") to round out capability coverage.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions — "Retrieve structures by UniProt ID", "download PDB/mmCIF files", "analyze confidence metrics (pLDDT, PAE)" — but omits capabilities present in the skill such as bulk proteome download and Google Cloud access, so coverage is not comprehensive.	4 / 5
Completeness	The "what" is clear (retrieve, download, analyze confidence metrics), but there is no explicit "Use when..." trigger clause; "for drug discovery and structural biology" is only a weak application-domain hint, so completeness is capped at 3.	3 / 5
Trigger Term Quality	Good natural keyword coverage including "AlphaFold", "protein structures", "UniProt", "PDB/mmCIF", "pLDDT", "PAE", "drug discovery", and "structural biology", but misses common variations like "AlphaFold DB" or "protein structure prediction/retrieval".	4 / 5
Distinctiveness Conflict Risk	Names a specific database (AlphaFold) with distinctive metric terms (pLDDT, PAE) and retrieval-oriented verbs, creating a clear niche with minimal overlap against generic structure-prediction skills.	5 / 5
	Total	16 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (525 lines); consider splitting into references/ and linking	Warning
metadata_version	'metadata.version' is missing	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	13 / 16 Passed

Repository: synthetic-sciences/openscience
Path: backend/cli/skills/databases/alphafold-database/SKILL.md
Commit: 4082a2e

Reviewed: 1 day 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.