skill-creator
Creates new AI agent skills following the Agent Skills spec. Trigger: When user asks to create a new skill, add agent instructions, or document patterns for AI.
75
70%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/skill-creator/SKILL.mdQuality
Discovery
67%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 adequately covers both what the skill does and when to use it, with an explicit trigger clause. However, it lacks specificity in concrete actions (e.g., generating YAML frontmatter, writing markdown) and could benefit from more natural trigger term variations that users would actually say. The distinctiveness is moderate but could be sharpened.
Suggestions
Add more specific concrete actions like 'generates YAML frontmatter, writes markdown skill instructions, defines trigger conditions and usage patterns'
Expand trigger terms to include natural variations like 'SKILL.md', 'skill file', 'teach Claude how to', 'skill template', 'agent capability'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain ('AI agent skills') and one action ('creates new'), but doesn't list specific concrete actions like 'generates YAML frontmatter, writes markdown instructions, defines trigger conditions'. The phrase 'following the Agent Skills spec' adds some specificity but remains somewhat vague about what concrete steps are involved. | 2 / 3 |
Completeness | Explicitly answers both 'what' (creates new AI agent skills following the Agent Skills spec) and 'when' (when user asks to create a new skill, add agent instructions, or document patterns for AI) with a clear 'Trigger:' clause providing explicit trigger guidance. | 3 / 3 |
Trigger Term Quality | Includes some relevant trigger terms like 'create a new skill', 'add agent instructions', and 'document patterns for AI', but misses common variations users might say such as 'SKILL.md', 'skill file', 'teach Claude', 'skill template', or 'agent behavior'. Coverage is partial. | 2 / 3 |
Distinctiveness Conflict Risk | The niche of 'AI agent skills' is fairly specific, but phrases like 'document patterns for AI' could overlap with documentation skills or general AI instruction skills. The term 'agent instructions' could also conflict with other agent-related skills. It's somewhat distinct but not fully carved out. | 2 / 3 |
Total | 9 / 12 Passed |
Implementation
72%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a solid, actionable skill for creating new agent skills. Its strengths are the concrete template, naming conventions table, and decision tree for assets vs references. Its main weaknesses are a lack of explicit workflow sequencing with validation steps (e.g., verifying frontmatter parsing or skill loading) and some minor verbosity in sections that explain concepts Claude would already understand.
Suggestions
Add an explicit numbered workflow (e.g., 1. Check if skill exists → 2. Create directory structure → 3. Write SKILL.md → 4. Validate frontmatter → 5. Register in AGENTS.md) with a validation checkpoint after writing the frontmatter.
Trim the 'When to Create a Skill' and 'Content Guidelines' sections — Claude can infer most of these heuristics; focus only on project-specific conventions that differ from defaults.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Generally efficient with good use of tables and decision trees, but some sections are slightly verbose (e.g., the 'When to Create a Skill' section explains somewhat obvious criteria, and the DO/DON'T section restates things Claude would infer). The template section is useful but could be tighter. | 2 / 3 |
Actionability | Provides a concrete SKILL.md template with exact frontmatter fields, a complete directory structure, naming convention examples, and a registration step with exact markdown syntax. The checklist and decision tree for assets/ vs references/ are directly actionable. | 3 / 3 |
Workflow Clarity | The checklist at the end provides a good sequence, and the overall flow (create structure → write SKILL.md → register in AGENTS.md) is implicit but not explicitly sequenced as numbered steps. There's no validation checkpoint — e.g., no step to verify the frontmatter parses correctly or that the skill loads properly after creation. | 2 / 3 |
Progressive Disclosure | Well-organized with clear sections using headers, tables, and code blocks. References to assets/ are clearly signaled. Content is appropriately structured for a single SKILL.md without unnecessary nesting or deeply chained references. | 3 / 3 |
Total | 10 / 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 |
|---|---|---|
allowed_tools_field | 'allowed-tools' contains unusual tool name(s) | Warning |
Total | 10 / 11 Passed | |
- Repository
- Gentleman-Programming/agent-teams-lite
- Commit
6901875
- 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.