Guide for creating effective, portable skills that extend Claude's capabilities with specialized knowledge, workflows, and tool integrations. Use when: (1) Creating a new skill from scratch, (2) Updating or improving an existing skill, (3) Structuring skill content for progressive disclosure, (4) Writing skill descriptions and triggers, (5) Packaging a skill for distribution. Triggers on: "create a skill", "build a skill", "new skill", "update this skill", "improve skill description", "skill structure", "skill triggers".
Install with Tessl CLI
npx tessl i github:ravnhq/ai-toolkit --skill agent-skill-creator97
Quality
100%
Does it follow best practices?
Impact
90%
1.16xAverage score across 3 eval scenarios
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 an excellent skill description that follows best practices. It uses third person voice, provides specific capabilities, includes explicit 'Use when:' and 'Triggers on:' clauses with comprehensive coverage of natural user phrases. The description clearly carves out a distinct niche for skill creation and management.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'creating effective, portable skills', 'extend Claude's capabilities with specialized knowledge, workflows, and tool integrations'. The numbered use cases add further specificity about creating, updating, structuring, writing descriptions, and packaging skills. | 3 / 3 |
Completeness | Clearly answers both what (guide for creating effective, portable skills with specialized knowledge/workflows/tool integrations) AND when (explicit 'Use when:' clause with 5 scenarios plus explicit 'Triggers on:' with 7 trigger phrases). | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural terms users would say, explicitly listed: 'create a skill', 'build a skill', 'new skill', 'update this skill', 'improve skill description', 'skill structure', 'skill triggers'. These are natural phrases users would actually type. | 3 / 3 |
Distinctiveness Conflict Risk | Very clear niche focused specifically on skill creation and management. The triggers are highly specific to skill-related tasks ('create a skill', 'skill structure', 'skill triggers') and unlikely to conflict with other skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
100%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is an exemplary skill that practices what it preaches. It demonstrates excellent token efficiency while remaining highly actionable, with clear workflow steps and appropriate progressive disclosure to reference files. The skill effectively teaches skill creation while serving as a model of good skill design.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is lean and efficient, assuming Claude's competence throughout. Every section earns its place with no unnecessary explanations of basic concepts. The structure table, workflow steps, and examples are all tightly written. | 3 / 3 |
Actionability | Provides concrete, executable guidance including exact folder structures, specific commands (scripts/init_skill.py, scripts/package_skill.py), clear workflow steps with entry/exit criteria, and specific examples of positive/negative triggers. | 3 / 3 |
Workflow Clarity | The 5-step workflow (Understand → Plan → Build → Test → Iterate) is clearly sequenced with explicit entry/exit conditions for each phase. Validation is addressed through references to validation-checklist.md and testing methodology, with clear feedback loops in the Iterate step. | 3 / 3 |
Progressive Disclosure | Excellent structure with a lean SKILL.md body that points to 6 well-organized reference files for detailed content. References are clearly signaled with descriptions of when to consult each one. The three-layer loading principle is both taught and practiced. | 3 / 3 |
Total | 12 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
metadata_field | 'metadata' should map string keys to string values | Warning |
Total | 10 / 11 Passed | |
Table of Contents
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.