solublempnn

Inverse-fold a backbone with SolubleMPNN — ProteinMPNN retrained on a soluble-PDB subset (Dauparas et al. 2022) — for sequences biased toward cytosolic expression and reduced aggregation. Reach for this skill when designs from vanilla ProteinMPNN are aggregating or going to inclusion bodies, when redesigning a membrane-adjacent fold for soluble expression, or when an E. coli expression screen is the next step.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

82%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 tight, actionable SKILL.md body built around a complete executable command, with the important gotchas (repo-relative imports, soluble weight versions, training-set vs host-solubility gap) surfaced rather than hidden. It earns its length on genuinely non-obvious domain context.

Dimension	Reasoning	Score
Conciseness	Mostly lean and competence-assuming with a complete runnable code block; a few prose sentences (e.g. the CPU-vs-GPU dispatching aside) could be trimmed without losing signal.	4 / 5
Actionability	Provides a copy-paste-ready bash block with all flags, calls out the load-bearing cd and the default --model_name gotcha, names the output path, and gives the LigandMPNN alternative with its exact flag, covering the common cases fully.	5 / 5
Workflow Clarity	The single run action is sequenced clearly with key gotchas surfaced and a fold-verification follow-up (boltz/esmfold2) recommended; it is not a multi-step destructive workflow, but the verification is framed as a follow-up rather than an inline checkpoint.	4 / 5
Progressive Disclosure	Cleanly sectioned (Running it, hydrophobic patches, training-set caveat) with one-level, well-signaled deferrals to sibling skills ("see proteinmpnn for the parsing quirks"); no bundle files exist, so structure is all inline and well-organized with only minor room to externalize.	4 / 5
	Total	17 / 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 well-crafted description that clearly states what SolubleMPNN does and gives three concrete, naturally-phrased trigger conditions. It is specific and distinct from vanilla ProteinMPNN, with only minor overlap risk against sibling MPNN skills.

Dimension	Reasoning	Score
Specificity	Names the concrete action ("Inverse-fold a backbone") plus the two outcome goals ("biased toward cytosolic expression and reduced aggregation"), giving several specific actions rather than a vague domain label, though coverage is essentially one core operation with its biases rather than a broad action list.	4 / 5
Completeness	Explicitly answers both what (inverse-fold with SolubleMPNN for soluble-expression-biased sequences) and when via a concrete "Reach for this skill when..." clause listing three trigger scenarios, matching the top anchor.	5 / 5
Trigger Term Quality	Strong natural phrases a domain user would actually say ("designs from vanilla ProteinMPNN are aggregating", "going to inclusion bodies", "E. coli expression screen"), but the set is jargon-heavy and lacks the synonym/extension breadth of the top anchor.	4 / 5
Distinctiveness Conflict Risk	Carves out a clear soluble-expression niche and contrasts against vanilla ProteinMPNN, but retains some overlap risk with the related proteinmpnn and ligandmpnn skills referenced in the body.	4 / 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/solublempnn/SKILL.md
Commit: 030c20b

Reviewed: 14 days 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.