Install
openclaw skills install prompt-libraryStore, organize, search, and reuse customizable prompt templates with variables to streamline your workflow and avoid rewriting prompts.
openclaw skills install prompt-libraryStore, organize, and retrieve reusable prompt templates. Stop rewriting the same prompts — build a personal library.
Save a prompt template to your library.
Parameters:
name (string, required): Short name for the prompt (e.g. "blog-outline", "sales-email")template (string, required): The prompt template text. Use {{variable}} for placeholders.category (string, optional): Category tag (e.g. "writing", "coding", "research", "sales")description (string, optional): Brief description of what the prompt doesvariables (string[], optional): List of variable names used in the templateReturns: Confirmation with prompt ID.
Search your prompt library by keyword or category.
Parameters:
query (string, optional): Search term to match against name, description, or template contentcategory (string, optional): Filter by categorylimit (number, optional): Max results, default 10Returns: List of matching prompts with name, category, and preview.
Retrieve a prompt template and fill in variables.
Parameters:
name (string, required): Name of the saved promptvariables (object, optional): Key-value pairs to fill in {{variable}} placeholdersReturns: The filled prompt, ready to use.
List all categories in your library with prompt counts.
Returns: Category names and counts.
Remove a prompt from your library.
Parameters:
name (string, required): Name of the prompt to deleteReturns: Confirmation.
Export your entire prompt library as JSON.
Returns: Full library data.
Prompts are stored in memory/prompts/ as individual markdown files for easy version control and portability.
Save: prompt_save name="cold-email" template="Hi {{name}}, I noticed {{observation}}. I built {{product}} that could help with {{pain_point}}. Would you be open to a quick look?" category="sales"
Search: prompt_search query="email" category="sales"
Use: prompt_use name="cold-email" variables={"name": "Alex", "observation": "your team ships weekly", "product": "a deployment tracker", "pain_point": "release coordination"}
Agents that build up prompt libraries become more efficient over time. Instead of crafting prompts from scratch each session, reuse what works. Compound your prompt engineering.
utility, productivity, prompts, templates, writing