skill-optimizer

Self-evolving skill optimization via SkillOpt-paper-grounded text-space optimizer.

Quality

61%

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/skill-optimizer/SKILL.md

Quality

Content

81%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.

The body is highly actionable and workflow-clear, with concrete commands, explicit validation gates, and feedback loops that fit a destructive/batch optimization skill. Its main weakness is moderate redundancy (the bundled-skill/held-out rule repeated three times) rather than padding.

Suggestions

State the bundled-skill in-place-mutation rule once (Iron Law) and reference it from the Decision tree and Anti-Patterns instead of restating it fully each time to save tokens.

If richer material exists, move the full walkthrough or rule-check vocabulary into a references/ file and link to it, raising progressive_disclosure structure.

Dimension	Reasoning	Score
Conciseness	The body is dense and assumes Claude's competence (citing D-codes like D17/D5/D12 without explanation) and avoids explaining basic concepts, but restates the bundled-skill/held-out rule three times (Iron Law, Decision tree, Anti-Patterns) and could be tightened; this fits 'Mostly efficient but includes some unnecessary explanation or could be tightened', above a 2 (no padded fluff) and below a 4 (the repetition is more than minor).	3 / 5
Actionability	Copy-paste-ready commands (e.g. 'gbrain skillopt X --bootstrap-from-skill', '--split 1:1:1'), a concrete JSONL line shape, exact flag semantics, exit codes, and a typed return contract fully cover the common cases, matching 'Fully executable; copy-paste ready code or commands; specific examples cover the common cases'.	5 / 5
Workflow Clarity	The pipeline and bootstrap sections give a clear sequenced flow with explicit validation checkpoints (median-of-3 + epsilon=0.05 gate, D_sel >= 5 preflight, cost preflight, sentinel-review loop) plus feedback loops (--resume, rejected.json, slow update), matching 'Clear sequence with explicit validation steps; feedback loops for error recovery; checklists for complex processes'.	5 / 5
Progressive Disclosure	No bundle files are present (references/scripts/assets are empty), so the score reflects the well-sectioned in-body structure with one clearly signaled one-level reference (the tutorial URL) and a Related skills list; it is below a 5 because content is not split into dedicated reference files, and above a 3 because structure and navigation are clear.	4 / 5
	Total	17 / 20 Passed

Description

41%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.

The description states a clear, distinctive purpose but is a single abstract sentence with no concrete actions, no natural trigger phrases, and no 'Use when' guidance. It reads as a tagline rather than a routing-ready description.

Suggestions

Add 2-3 concrete actions, e.g. 'Tunes a skill's body against a benchmark, validates candidates with median-of-3 gating, and writes a proposed.md for review.'

Append a 'Use when...' clause with natural trigger phrases such as 'Use when the user wants to make a skill better, tune it against a benchmark, or re-tune after switching target models.'

Include the user-facing term 'skillopt' and synonyms like 'tune', 'optimize', 'benchmark' so the description matches the listed triggers.

Dimension	Reasoning	Score
Specificity	Quotes "Self-evolving skill optimization" and "text-space optimizer" name the domain and one mechanism, but no concrete actions (e.g. tunes, validates, benchmarks) are listed, matching the anchor 'Names the domain but actions are minimal or generic'; it is above a 1 because it is not pure abstract fluff, but below a 3 which requires 1-2 concrete actions.	2 / 5
Completeness	It gives a clear 'what' (skill optimization via a text-space optimizer) but entirely lacks a 'when'/'Use when' clause, which the guideline caps at 3; it is above a 2 because the 'what' is not vague, and below a 4 which requires both what and when.	3 / 5
Trigger Term Quality	Only generic keywords like "skill optimization" appear with no natural phrases a user would say (e.g. "make the skill better", "tune the skill") and no synonyms or file extensions, fitting 'One or two generic keywords; missing the natural phrases users say'.	2 / 5
Distinctiveness Conflict Risk	The niche (paper-grounded skill optimization) is mostly distinct with only minor overlap risk against related meta-skills like skillify/skillpack-check; it is below a 5 because it supplies no distinct natural trigger phrases to fully disambiguate routing.	4 / 5
	Total	11 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: garrytan/gbrain
Path: skills/skill-optimizer/SKILL.md
Commit: f49ca56

Reviewed: 3 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.