Setup & Installation
openclaw skills install @mes28io/local-web-search-skill
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
npx clawhub install local-web-search-skill
What This Skill Does
Runs web searches locally by scraping DuckDuckGo HTML, returning structured JSON with results and per-source trust scores. No API key or paid service required. Designed as a fallback when standard web_search tools are unavailable.
Requires no signup, API key, or paid tier, making it a reliable fallback for any environment where external search APIs are unavailable.
When to use it
- Searching the web during local development with no API key
- Filtering search results by source trustworthiness before passing to a pipeline
- Querying for documentation or papers when web_search is missing or rate-limited
- Automating research tasks that feed results into downstream JSON processing
- Scoring and ranking sources to prioritize .gov, .edu, or authoritative domains
Example Workflow
Here's how your AI assistant might use this skill in practice.
INPUT
User asks: Find recent best practices for Python packaging
AGENT
- 1Run local_search.py with the query and --max 8
- 2Receive JSON response with results array and trust scores per result
- 3Filter results to high and medium trust tiers
- 4Extract title, url, and snippet fields for each qualifying result
OUTPUT
A filtered list of up to 8 results with titles, URLs, snippets, and trust tier labels, ready for further processing