openfold3

Structure prediction using OpenFold3, an open-weights PyTorch reproduction of AlphaFold3 from the AlQuraishi Lab. Use this skill when predicting protein/nucleic-acid/ligand complex structures with an Apache-2.0-licensed AF3 reimplementation.

Quality

85%

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

Quality

Content

88%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.

Excellent executable content: complete commands, a validated query-format spec, expected output structure with quality thresholds, and a thorough troubleshooting table. The only refinements are moving version-sensitive details out of the main flow and splitting troubleshooting/query detail into a reference file.

Dimension	Reasoning	Score
Conciseness	The body is dense and lean — prerequisites and parameter tables, no explanation of concepts Claude already knows. Not 5 because time-sensitive pins ("openfold3[cuequivariance]==0.4.1", checkpoint filename "of3-p2-155k.pt") appear outside any deprecated/old-patterns section.	4 / 5
Actionability	Fully executable throughout: pip install, huggingface-cli download, a complete run_openfold invocation, a valid query JSON covering protein and ligand cases, and concrete verify commands. Not 4 because commands are copy-paste ready and cover the common cases, including the offline/no-MSA variant.	5 / 5
Workflow Clarity	Clear install → weights → run → verify sequence with an explicit Verify section (expected sample counts), a 'What good output looks like' checklist, and an error→cause→fix troubleshooting table providing feedback loops. Not 4 because validation checkpoints and error recovery are explicit rather than implied.	5 / 5
Progressive Disclosure	Single-file skill with well-organized, clearly headed sections and no buried or nested references; the body is self-contained. Not 5 because at ~155 lines with no bundle files, the troubleshooting table and query-format detail are natural candidates for a one-level-deep reference file.	4 / 5
	Total	18 / 20 Passed

Description

82%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: it names a precise domain, states what the skill does, and gives an explicit 'Use this skill when' trigger clause in third person. The main gap is that it lists only one capability verb, so a reader learns the domain but not the breadth of actions the skill performs.

Dimension	Reasoning	Score
Specificity	"Structure prediction" is the sole concrete action; the domain is named precisely ("protein/nucleic-acid/ligand complex structures") but no secondary capabilities (MSA, templates, confidence analysis) are listed, matching 'names domain and 1-2 concrete actions'. Not 4 because 'several specific actions' are absent.	3 / 5
Completeness	Explicitly answers both: what ("Structure prediction using OpenFold3, an open-weights PyTorch reproduction of AlphaFold3") and when ("Use this skill when predicting protein/nucleic-acid/ligand complex structures") with concrete trigger phrases. The license qualifier in the when-clause is cosmetic, not a gap.	5 / 5
Trigger Term Quality	Natural phrases like "predicting", "protein", "nucleic-acid", "ligand", "complex structures", and "AlphaFold3"/"AF3" give good coverage. Not 5 because common synonyms such as "fold"/"folding", "biomolecular", or "PDB" are missing.	4 / 5
Distinctiveness Conflict Risk	"OpenFold3", "AlphaFold3", and molecule-type-specific triggers carve a clear niche with minimal overlap risk against any generic skill. Not 4 because it does not merely have 'minor overlap risk' — the tool-specific naming makes confusion unlikely.	5 / 5
	Total	17 / 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
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	13 / 16 Passed

Repository: UnicomAI/wanwu
Path: configs/microservice/bff-service/configs/agent-skills/claude-science/openfold3/SKILL.md
Commit: 4d4b150

Reviewed: 18 hours 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.