Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Web search API backed by the Presearch decentralized node network. Returns standard web results without user tracking or content filtering. Supports pagination, language, time range, and safe search parameters.
Unlike Google or Bing APIs, Presearch does not profile users or filter results, making it suitable for privacy-sensitive or research-oriented agent workflows.
When to use it
- Search the web from an AI agent without exposing queries to centralized logging
- Retrieve results on topics that major search engines filter or downrank
- Run time-scoped searches to find content published within the last day or week
- Fetch results in non-English languages for multilingual agent pipelines
- Page through large result sets programmatically across multiple requests
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: 'Find recent news about AI regulation'
- 1Constructs a GET request to the Presearch search endpoint with query 'AI regulation' and time set to 'week'
- 2Attaches the Bearer token in the Authorization header
- 3Receives a JSON response containing the standardResults array
- 4Extracts title, link, and description from each result object
- 5Returns the formatted list to the user with pagination metadata if more results exist
A list of recent web results with titles, URLs, and descriptions covering AI regulation from the past week
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.