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

83%

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

92%

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

The content is lean, highly actionable, and sequenced with validation checkpoints and troubleshooting feedback — strong on conciseness, actionability, and workflow clarity. Its only weakness is progressive disclosure: everything lives in one inline file with no external references.

Suggestions

Move the troubleshooting table and/or the detailed query JSON schema into a referenced file (e.g., references/troubleshooting.md) linked from the body, keeping SKILL.md as a leaner overview.

Optionally extract the full parameter table to a references/flags.md linked under 'Key parameters' to reduce inline density.

Dimension	Reasoning	Score
Conciseness	The body is dense and lean throughout — prerequisites table, copy-paste install/run commands, and a troubleshooting table with no explanatory padding of concepts Claude already knows; every section earns its tokens.	3 / 3
Actionability	Provides fully executable, copy-paste-ready commands (pip install, huggingface-cli download, run_openfold predict) plus a concrete query JSON example and a parameter table, all specific and complete rather than pseudocode.	3 / 3
Workflow Clarity	Sequences install → weights → running → verify with an explicit validation checkpoint ('Verify' section using grep/find) and a 'What good output looks like' checklist, plus a feedback-oriented troubleshooting table for error recovery.	3 / 3
Progressive Disclosure	The ~155-line body is well sectioned but entirely monolithic with no bundle files or one-level-deep references; content that could be split out (the troubleshooting table, detailed query JSON format) is inline, matching the 'content that should be separate is inline' anchor rather than the reference-driven level 3.	2 / 3
	Total	11 / 12 Passed

Description

75%

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 cleanly answers both 'what' and 'when' with a distinct niche and explicit trigger, but is somewhat single-action and leans technical in its trigger vocabulary.

Suggestions

Add a few more concrete action verbs (e.g., 'predict, score, and rank complex structures') to lift specificity to level 3.

Soften jargon with natural user phrasings like 'protein folding' or 'AlphaFold3 structure prediction' to broaden trigger-term coverage.

Dimension	Reasoning	Score
Specificity	Names the domain ('Structure prediction') and one concrete action ('predicting protein/nucleic-acid/ligand complex structures') but does not enumerate multiple distinct actions, so it sits at the 'names domain and some actions' anchor rather than the multi-action level 3.	2 / 3
Completeness	Explicitly states what ('Structure prediction using OpenFold3...') and when ('Use this skill when predicting protein/nucleic-acid/ligand complex structures...'), satisfying both halves with an explicit trigger clause.	3 / 3
Trigger Term Quality	Includes relevant natural terms ('protein', 'structure prediction', 'complex structures') but leans technical ('nucleic-acid/ligand complex', 'open-weights PyTorch reproduction of AlphaFold3') and misses common user variations like 'protein folding' or 'AlphaFold', placing it at 'some relevant keywords but missing common variations'.	2 / 3
Distinctiveness Conflict Risk	Targets a clear niche (OpenFold3 / AlphaFold3 complex structure prediction) with distinct triggers unlikely to fire for unrelated skills.	3 / 3
	Total	10 / 12 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: f618458

Reviewed: about 14 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.