Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
- Updated the project description for clarity and to fix a typo in "organization." - Added a new "When to Use" section to summarize scenarios where the rules apply.
What This Skill Does
Defines file placement conventions for skills that contain multiple features or generate output artifacts. Artifacts go in `data/`, feature scripts in `subskills/<feature>/`, and each feature folder can optionally include its own `SKILL.md`. Keeps the main skill root free of one-off scripts and generated files.
A shared layout convention means contributors always know where generated outputs and feature scripts live without reading each skill individually.
When to use it
- Deciding where to place a newly generated recommendation file
- Structuring a skill that has grown into multiple distinct features
- Adding a SKILL.md to document a specific subskill's behavior
- Auditing an existing skill folder for layout compliance
- Onboarding contributors to a skill's file organization conventions
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: 'I'm adding a new recommendation feature to my skill — where should the script and its output go?'
- 1Check if a `subskills/` directory exists in the skill root
- 2Create `subskills/<feature-name>/` for the new feature script
- 3Place the feature script inside that folder
- 4Create `data/` at the skill root for generated output artifacts
- 5Optionally add `SKILL.md` inside the feature folder to document its behavior
A clean skill layout with the feature script under `subskills/<feature-name>/` and generated outputs under `data/`, with no clutter in the skill root.