Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Version 1.2.1 - Added EXAMPLES.md with grouped command examples and edge-case workflows. - SKILL.md updated for clarity and conciseness, with a reorganized structure and quick-reference sections. - Improved documentation on safety: now warns not to use sensitive files as `SCHEMA_FILE` and lists file blocks. - Enhanced explanations on API variable usage, category/domain constraints, and agent decision rules for search types. - Emphasized common pitfalls to help avoid errors and improve reliability.
What This Skill Does
Exa is a search and research API that runs web, neural, code, and category-filtered searches alongside full page content extraction. It also executes async multi-step research tasks with optional structured output schemas.
Combining neural search, content extraction, and async research in one API removes the need to chain separate tools for search-then-read workflows.
When to use it
- Searching recent news articles filtered by domain and date range
- Extracting full text and subpages from a list of URLs
- Finding code documentation and examples for a specific programming library
- Running async deep research on a company and getting structured JSON output
- Searching LinkedIn profiles when researching specific people
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: 'Find recent news about OpenAI funding rounds from the past month'
- 1Sets CATEGORY=news and SINCE to one month ago
- 2Runs search.sh with the query to retrieve matching articles
- 3Collects top result URLs from the search response
- 4Runs content.sh on those URLs to extract full page text
A set of recent news articles about OpenAI funding with full extracted text ready for summarization
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.