Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Runs a search query across Bing, Yahoo, Startpage, AOL, and Ask simultaneously using async-search-scraper. Returns combined results as JSON, CSV, or plain text with deduplication options. Supports proxies and configurable page depth per engine.
Querying five engines at once in a single script run yields broader coverage than any single engine without requiring separate API keys or browser automation.
When to use it
- Aggregating search results from multiple engines in one pass
- Cross-referencing a topic across engines to surface obscure results
- Collecting URLs for a domain or topic for further scraping
- Running privacy-conscious searches via a SOCKS proxy
- Deduplicating results by domain to get a broader source spread
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: find recent articles about prompt injection attacks
- 1Run web_multi_search.py with the query and --unique-urls flag
- 2Capture JSON output
- 3Parse results to extract link, title, and text fields
- 4Filter or rank entries by relevance to the query
- 5Return top links and snippets to the user
A ranked list of URLs and page snippets from Bing, Yahoo, Startpage, AOL, and Ask covering the topic