openfold3

Run or plan OpenFold3-style structure prediction workflows. Use when a task asks for open protein or complex prediction, setup, model comparison, or reproducibility around OpenFold-family outputs.

Quality

78%

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 ./skills/openfold3/SKILL.md

Quality

Content

72%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.

A compact, well-structured workflow overview that respects the token budget and organizes content cleanly. Its main weakness is actionability and feedback-loop completeness: it tells Claude what to do and what to save but not how to execute or recover when verification fails.

Suggestions

Add one or two executable snippets or concrete commands (e.g., a sample run command, a validation/lockfile command) so the 'how' is as concrete as the 'what'.

Turn step 2's 'Verify ... before running' into an explicit feedback loop: 'If verification fails, fix the missing database/checkpoint/GPU route and re-verify before proceeding.'

Specify how reproducibility gaps from step 5 should be resolved or surfaced (e.g., halt and report, or proceed with a documented approximation) so the recovery path is unambiguous.

Dimension	Reasoning	Score
Conciseness	Lean ~12-line body that assumes Claude's competence (no explanation of FASTA, MSA, PDB, checkpoints, or GPU routes) with every line carrying an actionable instruction and zero padding, matching the lean-and-efficient anchor.	5 / 5
Actionability	Names concrete artifacts to handle (manifest, environment lockfile, confidence files, runtime metadata) and clear per-step outcomes, but provides no executable commands or code showing how to normalize, verify, or run, leaving key execution details missing.	3 / 5
Workflow Clarity	A clear five-step sequence with an explicit pre-run verification checkpoint (step 2) and a reproducibility-gap reporting step (step 5), but it lacks a validate->fix->retry feedback loop for a heavy/batch GPU compute workflow, which caps it at 3 per the rubric.	3 / 5
Progressive Disclosure	An under-50-line single-purpose skill with a heading, a one-line intro, a numbered workflow, and a closing guardrail; no external references are needed and the organization is clean, satisfying the simple-skills exception for a top score.	5 / 5
	Total	16 / 20 Passed

Description

83%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, concise description that clearly pairs a concrete capability statement with an explicit 'Use when' trigger clause scoped to the OpenFold-family niche. It is highly complete and reasonably specific, with only minor overlap risk against adjacent structure-prediction skills.

Dimension	Reasoning	Score
Specificity	Names the domain and several concrete actions ('Run or plan OpenFold3-style structure prediction workflows', 'setup, model comparison, or reproducibility'), though 'model comparison' and 'reproducibility' read more as goals than granular operations, leaving minor coverage gaps versus the comprehensive anchor 5.	4 / 5
Completeness	Explicitly states what it does ('Run or plan OpenFold3-style structure prediction workflows') and pairs it with a concrete 'Use when a task asks for...' trigger clause enumerating multiple use cases, matching the both-what-and-when anchor.	5 / 5
Trigger Term Quality	Good natural-term coverage ('open protein or complex prediction', 'model comparison', 'reproducibility', 'OpenFold-family', 'OpenFold3') that a bioinformatics user would say, but a few synonyms (e.g., 'structure prediction', 'AlphaFold comparison') and concrete file-type triggers are absent.	4 / 5
Distinctiveness Conflict Risk	The OpenFold-family/OpenFold3-style niche is clearly distinct with scoped triggers, but broad phrases like 'model comparison' and 'protein or complex prediction' create minor overlap risk with closely related AlphaFold/ESMFold structure-prediction skills.	4 / 5
	Total	17 / 20 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: companion-inc/feynman
Path: skills/openfold3/SKILL.md
Commit: 95c51d9

Reviewed: about 9 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.