Create or update ZeptoClaw skills. Use when designing, structuring, or authoring new agent skills.
Guide users through designing and building ZeptoClaw skills. Follow these six steps in order — do not skip ahead.
Before designing anything, ask the user for 3-5 real queries or tasks the skill should handle. Collect actual example inputs and expected behaviors. Do not proceed until you have concrete use cases — vague ideas produce vague skills.
For each example, ask: "What would need to be looked up, computed, or re-done each time without this skill?" The answers become the skill's content. Discard anything the agent already knows how to do without help.
Categories of reusable content:
If nothing survives this filter, the user may not need a skill — say so.
Pick the structural pattern that best fits the content:
| Pattern | When to use | Example |
|---|---|---|
| Workflow | Ordered multi-step process | deploy-docker, release-notes |
| Task-based | Independent tasks, any order | shopee, github |
| Reference | Lookup information | weather, api-cheatsheet |
| Capabilities | Agent behaviors/personas | code-reviewer, translator |
Then assign a freedom level to each piece of content:
Match specificity to fragility — only lock down what breaks when done differently.
Create the skill directory and file:
zeptoclaw skills create <skill-name>
This creates ~/.zeptoclaw/skills/<skill-name>/SKILL.md with a starter template.
If the skill needs bundled resources, create subdirectories:
scripts/ — executable code the agent runs (deterministic, token-efficient)references/ — documentation loaded into context on demandassets/ — output files (templates, icons) never loaded into contextNaming rules: Lowercase letters, digits, and hyphens only. Max 64 characters. No leading, trailing, or consecutive hyphens. Prefer verb-led names: deploy-docker, track-expenses, format-invoice.
Every SKILL.md starts with YAML frontmatter:
---