A privacy-focused, zero-cost web search tool that uses DuckDuckGo scraping to provide high-quality search results without requiring external API keys.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install local-web-search-skill
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install local-web-search-skill using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Local Web Search?
Local Web Search is a robust utility designed for developers who need to integrate real-time information retrieval into their workflows without the overhead of paid search APIs. By leveraging DuckDuckGo HTML scraping, this tool provides a reliable way to query the web while maintaining high security standards and data privacy. It is an essential addition to the Openclaw Skills ecosystem for those building cost-effective AI agents.
The tool includes built-in source trust scoring to help developers and agents evaluate the authority of the returned results. By distinguishing between official documentation, reputable publications, and user-generated content, it ensures that subsequent reasoning steps are based on the most reliable data available.
Local Web Search Use Cases
- Retrieving real-time data for AI agent pipelines without incurring API costs.
- Replacing the default web search tool when API keys are missing or expired.
- Conducting automated research with built-in source reliability ranking.
- Building privacy-preserving search workflows that avoid tracking by major search providers.
How Local Web Search Works
- The user or AI agent executes the Python script with a specific search query and an optional result limit.
- The script performs an outbound HTTPS GET request to the DuckDuckGo search endpoint.
- A specialized parser extracts result titles, URLs, and snippets from the HTML response.
- The system calculates a trust score for each result based on domain authority tiers.
- A formatted JSON object is returned, containing the ranked results and metadata for programmatic processing.
Local Web Search Setup
Ensure you have Python installed in your environment. You can execute the search script directly from your terminal within the Openclaw Skills directory structure:
./skills/local-web-search/scripts/local_search.py "your search query" --max 8
Local Web Search Data Schema & Taxonomy
The skill returns a structured JSON object designed for easy parsing by AI models. The schema includes:
| Field | Description |
|---|---|
query |
The original search query processed. |
count |
Total number of results retrieved. |
results |
An array containing title, url, snippet, and trust objects. |
trust.tier |
Ranking of the source (high, medium, or low). |
security |
Metadata confirming no environment exfiltration occurred. |
Local Web Search Advanced Features
- Automated Trust Scoring: Categorizes results into tiers like high (docs/papers) and low (community platforms) to improve agent reasoning.
- Exponential Backoff: Includes built-in retry logic with jitter to handle transient network errors or rate limiting gracefully.
- Privacy-First Design: Features no external writes and strictly limited outbound HTTPS GET requests to ensure secure operation within Openclaw Skills.
- JSON Pipeline Compatibility: Optimized output format specifically designed for multi-agent workflows and automated data extraction.
SKILL.md
Loading
Related Openclaw Skills
Bulk TikTok Downloader
A command-line utility for batch downloading TikTok videos from a text file using yt-dlp for efficient media archival.
SerpAPI MCP Search
A sophisticated search integration that connects AI agents to SerpAPI via the Model Context Protocol for real-time, structured web data.
Hello World Skill
A fundamental greeting skill for AI agents that provides personalized, multilingual messages with dynamic date and time integration.
Live-Evo: Self-Evolving AI Memory System
A dynamic learning framework that enables AI agents to accumulate experience, verify outcomes, and self-correct through a persistent memory database.
Minimalist UI Design System
A specialized design system and implementation playbook for creating clean, high-density interfaces inspired by Notion and OpenAI aesthetics.
Bot Debate
A specialized skill enabling AI agents to autonomously compete in structured, multi-turn debates via a standardized RESTful HTTP interface.
Featured*
