skill-creator
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.
65
46%
Does it follow best practices?
Impact
98%
1.42xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./data/skills-md/0xdarkmatter/claude-mods/skill-creator/SKILL.mdQuality
Discovery
57%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 successfully communicates when to use the skill with an explicit 'Use when' clause, but falls short on specificity—it doesn't describe concrete actions like generating frontmatter, structuring markdown, or validating skill files. The trigger terms are adequate but could include more natural variations users might say.
Suggestions
Add specific concrete actions the skill performs, e.g., 'Generates YAML frontmatter, structures markdown instructions, defines trigger conditions, and validates skill file format.'
Include additional trigger term variations such as 'SKILL.md', 'skill file', 'write a skill', 'custom skill template' to improve matching coverage.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description says 'creating effective skills' and 'extends Claude's capabilities with specialized knowledge, workflows, or tool integrations' but doesn't list concrete actions like 'generates YAML frontmatter, writes markdown instructions, validates skill structure.' The language is abstract and vague about what it actually does. | 1 / 3 |
Completeness | It answers both 'what' (guide for creating effective skills) and 'when' ('should be used when users want to create a new skill or update an existing skill'). The 'when' clause is explicit with clear trigger conditions. | 3 / 3 |
Trigger Term Quality | Includes some relevant terms like 'create a new skill', 'update an existing skill', and 'skill' which users would naturally say. However, it misses variations like 'SKILL.md', 'skill file', 'write a skill', 'skill template', or 'custom instruction'. | 2 / 3 |
Distinctiveness Conflict Risk | The term 'skill' is somewhat specific to this domain, but 'extends Claude's capabilities with specialized knowledge, workflows, or tool integrations' is broad enough to potentially overlap with other meta-skills or configuration-related skills. The niche is identifiable but not sharply delineated. | 2 / 3 |
Total | 8 / 12 Passed |
Implementation
35%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is comprehensive in coverage but suffers significantly from verbosity—it explains many concepts Claude already understands (what skills are, what PDFs are, what scripts do) and includes extensive meta-commentary that inflates token cost without proportional value. The workflow is reasonably clear with a 6-step process, but the skill would benefit greatly from practicing the conciseness principles it preaches. The content about progressive disclosure patterns and resource types could itself be moved to reference files.
Suggestions
Cut the 'About Skills' and 'What Skills Provide' sections entirely or reduce to 2-3 lines—Claude already knows what skills are since it's reading one. Move the detailed progressive disclosure patterns and bundled resource explanations to a reference file.
Remove explanatory prose throughout (e.g., 'Think of them as onboarding guides', 'Think of Claude as exploring a path') and replace with terse, imperative instructions. Target reducing the body by at least 50%.
Add explicit validation checkpoints in Step 4 (Edit the Skill), such as verifying frontmatter format, checking line count limits, and confirming all referenced files exist before proceeding to packaging.
Practice progressive disclosure by moving the detailed 'Anatomy of a Skill' section, resource type explanations, and progressive disclosure patterns into a references/skill-design.md file, keeping only a brief summary and link in SKILL.md.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~400+ lines, with extensive explanations of concepts Claude already knows (what skills are, what PDFs are, what scripts/references/assets mean). Sections like 'What Skills Provide' and 'About Skills' explain basic concepts that don't earn their token cost. Many paragraphs could be reduced to bullet points or eliminated entirely. | 1 / 3 |
Actionability | The skill provides some concrete guidance (e.g., specific bash commands for init_skill.py and package_skill.py, frontmatter examples, directory structures), but much of the content is meta-guidance and principles rather than executable instructions. The actual skill creation steps mix actionable commands with abstract advice. | 2 / 3 |
Workflow Clarity | The 6-step process is clearly sequenced and numbered, which is good. However, validation is only mentioned in the packaging step (Step 5), and there are no explicit checkpoints or feedback loops during the critical editing phase (Step 4). The iteration step (Step 6) is vague and lacks concrete validation criteria for determining if a skill is working well. | 2 / 3 |
Progressive Disclosure | The skill references two external files (references/workflows.md and references/output-patterns.md) and external URLs, which is good. However, the SKILL.md itself is monolithic and contains extensive inline content about progressive disclosure patterns, bundled resource types, and design principles that could be split into reference files. The irony is that a skill about progressive disclosure doesn't practice it well enough itself. | 2 / 3 |
Total | 7 / 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
- NeverSight/skills_feed
- Commit
3ae408c
- 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.