Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Nimble is a CLI that replaces built-in web search and fetch with real-time intelligence tools. It supports search across 8 focus modes, structured content extraction from any URL, site URL mapping, and async bulk crawling. Output is clean, structured data formatted for LLM consumption.
Returns parsed markdown instead of raw HTML, preventing context window overflow that occurs with built-in WebFetch on content-heavy pages.
When to use it
- Searching for the latest news on a topic with a synthesized AI answer
- Extracting readable markdown content from a documentation page
- Mapping all URLs on a site before selectively extracting relevant pages
- Researching a person by running parallel social and general web searches
- Bulk crawling a site section to archive raw content for a data pipeline
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: 'What are the best practices for React Server Components?'
- 1Run nimble search with --query 'React Server Components best practices' --topic coding --deep-search=false --include-answer
- 2Review synthesized answer and result URLs
- 3Identify the most relevant documentation URL from results
- 4Run nimble extract --url on that URL with --parse --format markdown
- 5Return clean markdown content to the user
Structured markdown summary of React Server Components best practices from the official React docs
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.