ds-optimize

Use when an algorithm-first quest should manage candidate briefs, optimization frontier, branch promotion, or fusion-aware search instead of the paper-oriented default loop.

Quality

70%

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

Fix and improve this skill with Tessl

tessl review fix ./skills/ds-optimize/SKILL.md

Quality

Content

55%

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

The content is highly actionable and well-sequenced with strong validation feedback loops. Its serious weaknesses are conciseness — the same templates and playbooks are stated twice across the body and the inlined appendix — and progressive disclosure, since a 1600-line skill that inlines all reference material is effectively a monolith rather than an overview pointing to one-level-deep references.

Suggestions

De-duplicate the body and the integrated appendix: keep each template/playbook once and have the body reference it, cutting roughly half the appendix.

Move the templates and playbooks into actual files under references/ (e.g. references/brief-shaping-playbook.md) and point to them one level deep instead of inlining them, restoring progressive disclosure.

Tighten repetitive rule restatements (the submit_idea candidate-vs-line distinction is explained in at least five sections) into a single canonical statement referenced where needed.

Dimension	Reasoning	Score
Conciseness	At ~1649 lines the body is highly verbose and repeats itself: the brief-shaping playbook, candidate-ranking template, and method-brief template each appear in the main body and again verbatim in the integrated appendix, padding the context with material Claude does not need stated twice.	1 / 3
Actionability	It provides concrete, copy-paste-ready MCP calls (e.g. `artifact.submit_idea(mode='create', submission_mode='candidate', ...)`), explicit field lists, status enumerations, and filled-in templates, giving fully executable guidance rather than abstract direction.	3 / 3
Workflow Clarity	The Required MCP-driven workflow is a numbered 7-step sequence with explicit state-recovery checkpoints ("Recover the optimization state first", "Do not create new candidates before the frontier... is checked") and route-from-evidence feedback loops, including a plateau/stall recovery path.	3 / 3
Progressive Disclosure	No references/ scripts/ or assets/ bundle exists; instead the appendix inlines the former reference files inline ("This appendix inlines the former optimize/references/*.md"), producing a monolithic wall of text with no real file separation or navigation despite the body being large enough to warrant it.	1 / 3
	Total	8 / 12 Passed

Description

85%

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 is specific, complete with an explicit "Use when" trigger, and clearly distinguished from paper-oriented work. Its main weakness is trigger-term quality: the natural-language keywords lean on internal DeepScientist jargon that users are unlikely to actually say.

Suggestions

Add user-natural phrasing to the trigger clause, e.g. "when the user asks to improve an algorithm's result, tune an experiment, or rank competing approaches".

Mirror common surface terms (e.g. "optimization", "candidate search", "experiment tuning") alongside the internal vocabulary so the description matches what a user would say.

Dimension	Reasoning	Score
Specificity	The description names multiple concrete actions — "manage candidate briefs, optimization frontier, branch promotion, or fusion-aware search" — matching the anchor that lists several specific capabilities rather than vague single-domain phrasing.	3 / 3
Completeness	It explicitly answers both what (the four management actions) and when ("Use when an algorithm-first quest should... instead of the paper-oriented default loop"), with an explicit trigger clause.	3 / 3
Trigger Term Quality	It includes a "Use when" clause and a few relevant terms, but the keywords ("optimization frontier", "fusion-aware search", "branch promotion") are internal jargon a user would rarely say; common natural variations are missing, so it is not a score-3 coverage of user language.	2 / 3
Distinctiveness Conflict Risk	The niche is narrow and contrasted explicitly against "the paper-oriented default loop", giving it a distinct trigger unlikely to fire for unrelated skills.	3 / 3
	Total	11 / 12 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (1650 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: OpenLAIR/dr-claw
Path: skills/ds-optimize/SKILL.md
Commit: 6da7f7c

Reviewed: about 8 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.