Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Add ddgs-search CLI wrapper script that fixes raw ddgs pagination bug (input() → Aborted + exit 1 in non-TTY). Wrapper outputs clean JSON, exit 0. Updated install.py to auto-install wrapper to ~/.local/bin. Updated SKILL.md docs.
What This Skill Does
A CLI-based web search tool that queries Google, Bing, DuckDuckGo, Brave, Yandex, Yahoo, Wikipedia, and arXiv from a single interface. No API keys or accounts required. Outputs JSON compatible with web-search-plus, making it a drop-in search backend for other skills.
Covers 8 search engines and arXiv in one install with no API keys, billing, or rate limits.
When to use it
- Searching the web without setting up API credentials
- Finding recent academic papers on a specific ML or AI topic
- Switching search engines to compare result sets
- Plugging in a search backend for other installed skills
- Querying arXiv by title, abstract, or category field
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Find recent papers on 3D Gaussian splatting
- 1Run arxiv_search.py with query '3D gaussian splatting' and max results set to 10
- 2Parse returned JSON for title, authors, abstract, categories, and publication date
- 3Filter or re-sort by relevance if the default date ordering is not useful
- 4Return structured paper list to the user
10 arXiv papers with titles, authors, categories, abstracts, and publication dates