skill-creator
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, 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.
85
81%
Does it follow best practices?
Impact
88%
1.87xAverage score across 3 eval scenarios
Advisory
Suggest reviewing before use
Quality
Discovery
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.
This is a strong description that clearly defines a specific domain (skill management and optimization), lists concrete actions, and includes an explicit 'Use when...' clause with natural trigger terms. The description is well-structured, concise, and covers both the capabilities and the activation conditions comprehensively. It occupies a distinct niche that is unlikely to conflict with other skills.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: create new skills, modify/improve existing skills, measure skill performance, run evals, benchmark with variance analysis, and optimize descriptions for triggering accuracy. | 3 / 3 |
Completeness | Clearly answers both 'what' (create, modify, improve, measure skills) and 'when' with an explicit 'Use when...' clause listing specific trigger scenarios like creating from scratch, editing, running evals, benchmarking, and optimizing descriptions. | 3 / 3 |
Trigger Term Quality | Includes strong natural trigger terms users would say: 'create a skill', 'edit', 'optimize', 'run evals', 'test a skill', 'benchmark', 'skill performance', 'triggering accuracy', 'description'. These cover a good range of how users would phrase requests about skill management. | 3 / 3 |
Distinctiveness Conflict Risk | The domain of skill creation, editing, evaluation, and optimization is a clear niche. Terms like 'skill', 'evals', 'variance analysis', 'triggering accuracy' are highly specific and unlikely to conflict with other skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
62%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a highly actionable and well-structured skill with excellent workflow clarity, providing concrete commands, JSON schemas, and step-by-step processes for creating and iterating on skills. However, it suffers significantly from verbosity — conversational asides, repeated instructions (the core loop appears 3+ times), motivational commentary, and explanations of things Claude already understands inflate the token cost substantially. The content organization would benefit from moving platform-specific sections and the description optimization workflow into separate reference files.
Suggestions
Cut conversational filler ('Cool? Cool.', 'Sorry in advance but...', 'Good luck!'), motivational language ('billions a year in economic value'), and repeated restatements of the core loop — the loop is stated at least 3 times and once is sufficient.
Move platform-specific sections (Claude.ai instructions, Cowork instructions) into separate reference files (e.g., references/claude-ai.md, references/cowork.md) and reference them from SKILL.md with one-line pointers.
Remove explanations of concepts Claude already knows — e.g., why LLMs benefit from understanding 'why', what progressive disclosure means conceptually, the trend of non-technical users using terminals.
Consider moving the Description Optimization section into its own reference file (e.g., references/description-optimization.md) since it's a distinct sub-workflow that adds ~100 lines to the main skill.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~500+ lines with significant conversational padding ('Cool? Cool.'), explanations of concepts Claude already knows (what skills are, how LLMs work), motivational language ('we are trying to create billions a year in economic value here!'), and repeated instructions (the core loop is stated 3+ times). Much of the content could be cut by 40-50% without losing information. | 1 / 3 |
Actionability | Despite verbosity, the skill provides highly concrete, executable guidance: specific CLI commands, exact JSON schemas, file path conventions, code blocks for spawning subagents, grading, aggregation, and viewer generation. The step-by-step processes include copy-paste ready commands and specific file formats. | 3 / 3 |
Workflow Clarity | The multi-step workflow is clearly sequenced with explicit steps (Step 1 through Step 5), validation checkpoints (grade runs, aggregate benchmarks, analyst pass before showing to user), feedback loops (iterate until user is satisfied), and clear error recovery patterns. The iteration loop is well-defined with explicit stopping criteria. | 3 / 3 |
Progressive Disclosure | The skill references external files appropriately (agents/grader.md, agents/comparator.md, agents/analyzer.md, references/schemas.md) with clear guidance on when to read them. However, the SKILL.md body itself is monolithic and contains substantial content that could be split into reference files — the description optimization section, Claude.ai-specific instructions, and Cowork-specific instructions could each be separate files, keeping the main skill leaner. | 2 / 3 |
Total | 9 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- anthropics/skills
- Commit
f458cee
- Reviewed
Table of Contents
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.