skill-creator

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, update or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

Quality

88%

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

The canonical home for this skill is skill-creator in anthropics/skills

Quality

Content

77%

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 with a well-sequenced, validation-checkpointed workflow, but it is over-long and chatty and — more seriously — references several bundle files (agents/*.md, eval-viewer/generate_review.py) that are absent from the skill, which hurts navigation.

Suggestions

Add the missing agents/ directory (grader.md, comparator.md, analyzer.md) and eval-viewer/generate_review.py that the body repeatedly references, or remove those references, so the one-level-deep navigation actually resolves.

Trim conversational filler (e.g. 'Cool? Cool.', the plumbers/parents anecdote, the all-caps aside) and the repeated restatements of the core loop to bring the body closer to the 500-line lean ideal.

Consider moving the long Claude.ai-specific and Cowork-specific instruction blocks into separate reference files so the main SKILL.md body stays a tighter overview.

Dimension	Reasoning	Score
Conciseness	The 481-line body is padded with conversational filler ('Cool? Cool.', the plumbers/parents anecdote, 'Sorry in advance but I'm gonna go all caps here') and repeats the core loop several times, matching the level-2 'mostly efficient but includes some unnecessary explanation or could be tightened' anchor rather than the level-3 lean/efficient bar. It is not level 1 because substantial portions (real commands, JSON field names, file paths) genuinely earn their tokens.	2 / 3
Actionability	Provides fully executable commands (`python -m scripts.aggregate_benchmark <workspace>/iteration-N --skill-name <name>`, the nohup generate_review.py invocation), exact JSON field names ('text', 'passed', 'evidence'), and concrete file paths, hitting the level-3 'fully executable code/commands; specific examples; copy-paste ready' anchor.	3 / 3
Workflow Clarity	The test-and-evaluate section is an explicitly sequenced Step 1–5 flow with validation checkpoints and feedback loops ('If errors: fix and re-validate', 'Only when valid'), plus a clearly staged iteration loop, matching the level-3 'clear sequence with explicit validation steps; feedback loops for error recovery' anchor.	3 / 3
Progressive Disclosure	Real one-level references are well signaled (`See references/schemas.md`, `assets/eval_review.html`, `scripts/`), but the body repeatedly points to `agents/grader.md`, `agents/comparator.md`, `agents/analyzer.md`, and `eval-viewer/generate_review.py` — files whose directories do not exist in the bundle. These dangling references are a navigation/organization defect, placing it at level-2 'some structure but could be better organized' rather than level-3.	2 / 3
	Total	10 / 12 Passed

Description

100%

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 strong across all four dimensions: it names multiple concrete actions, includes a clear explicit 'Use when' trigger clause, and occupies a distinctive niche. It is one of the better-structured descriptions in the reference set.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'Create new skills, modify and improve existing skills, and measure skill performance' plus 'update or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description' — matching the level-3 anchor that asks for several specific concrete actions rather than a single domain.	3 / 3
Completeness	Explicitly answers both what ('Create new skills, modify and improve existing skills, and measure skill performance') and when via an explicit 'Use when users want to create a skill from scratch...' clause, which is the level-3 'clearly answers both what AND when with explicit triggers' anchor.	3 / 3
Trigger Term Quality	Covers natural phrases a user would actually say — 'create a skill from scratch', 'run evals to test a skill', 'benchmark skill performance', 'optimize a skill's description for better triggering accuracy' — giving good coverage of natural terms rather than just technical jargon. Not the level-2 'some relevant keywords but missing common variations' anchor because it spans several distinct intent phrasings.	3 / 3
Distinctiveness Conflict Risk	Occupies a clear niche — skill authoring, eval running, benchmarking, and description optimization — with distinct triggers unlikely to fire for unrelated skills, matching the level-3 'clear niche with distinct triggers; unlikely to conflict' anchor.	3 / 3
	Total	12 / 12 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: netease-youdao/LobsterAI
Path: SKILLs/skill-creator/SKILL.md
Commit: 2921c1e

Reviewed: about 11 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.