ligandmpnn

Design protein sequences around ligand or small-molecule contexts with LigandMPNN-style workflows. Use when a task asks for ligand-aware protein design, residue redesign, constraints, or design ranking.

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

78%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, well-structured workflow that respects token budget and sequences verification appropriately, but it stays at the level of a checklist rather than giving executable commands, tool syntax, or quantitative filter thresholds.

Suggestions

Add a concrete invocation example or command skeleton for the LigandMPNN-style run (e.g. sample CLI flags or a minimal script snippet) so step 2-3 are executable rather than descriptive.

Specify quantitative or checkable filter thresholds for step 4 (e.g. minimum sequence identity for diversity, confidence cutoff, contact-distance criteria) so filtering is reproducible.

Add an explicit failure feedback loop after step 2 and step 5 (what to do if the route/model is unavailable or the independent check flags a design) to close the validation loop.

Dimension	Reasoning	Score
Conciseness	Lean and efficient: a short overview plus a five-step workflow with no padding and no explanation of concepts Claude already knows; every token earns its place.	5 / 5
Actionability	Names concrete artifacts and filter criteria ('generated FASTA, and score tables', 'constraint satisfaction, sequence diversity, known motifs, predicted structure confidence, and ligand-contact plausibility') but provides no executable commands, tool invocation syntax, or quantitative thresholds, leaving the 'how' underspecified.	3 / 5
Workflow Clarity	Clear five-step sequence with explicit checkpoints (step 2 'Verify the execution route and model availability before running'; step 5 independent check 'before presenting them as candidates'), but no explicit failure-handling feedback loop describing what to do when verification fails.	4 / 5
Progressive Disclosure	Under 50 lines, single-purpose, with no bundle files present and no need for external references; the numbered Workflow list is well-organized and self-contained.	5 / 5
	Total	17 / 20 Passed

Description

87%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 concise, well-targeted description that answers both what the skill does and when to use it, in third person with concrete trigger phrases. Minor gains available by adding synonym terms and a couple more action verbs.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions — 'Design protein sequences', 'residue redesign, constraints, or design ranking' — naming the domain with multiple specific capabilities, though coverage stops short of being fully comprehensive (no scoring/filtering verbs).	4 / 5
Completeness	Explicitly states the 'what' ('Design protein sequences around ligand or small-molecule contexts with LigandMPNN-style workflows') and a concrete 'Use when…' trigger clause covering multiple scenarios.	5 / 5
Trigger Term Quality	Includes natural terms a user would say ('ligand-aware protein design', 'residue redesign', 'constraints', 'design ranking') but lacks synonyms (e.g. 'small molecule', 'binding site', 'mutation') and file extensions.	4 / 5
Distinctiveness Conflict Risk	Targets a clear niche ('LigandMPNN-style workflows', 'ligand-aware protein design') with distinct triggers and minimal overlap risk against other skills.	5 / 5
	Total	18 / 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/ligandmpnn/SKILL.md
Commit: 95c51d9

Reviewed: about 10 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.