A privacy-focused metasearch skill that aggregates results from over 70 engines using a self-hosted SearXNG instance.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install searxng-self-hosted
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 searxng-self-hosted 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 SearXNG Search Skill?
The SearXNG search skill is a powerful addition to the Openclaw Skills ecosystem, enabling developers to perform privacy-respecting web searches without relying on a single provider. By acting as a metasearch engine, it pulls data from Google, DuckDuckGo, Brave, and dozens of others, providing a neutral and comprehensive view of the web.
This skill is designed for users who value data sovereignty and require a reliable API for web research. Because it relies on a self-hosted instance, there are no tracking scripts, ads, or third-party data collection, making it an ideal choice for secure Openclaw Skills integrations. It serves as a bridge between your local environment and the global web while maintaining strict privacy standards.
SearXNG Search Skill Use Cases
- Conducting automated research across multiple search engines simultaneously.
- Gathering technical documentation from specific categories like IT or Science.
- Integrating private search capabilities into custom developer tools within Openclaw Skills.
- Performing localized searches across different languages and regions without user tracking.
- Developing automated news aggregators or monitoring tools.
How SearXNG Search Skill Works
- The skill connects to a self-hosted SearXNG instance via a configurable URL set in the environment variables.
- It sends search queries with optional parameters for specific categories, language preferences, or pagination.
- SearXNG aggregates results from over 70 different search engines in real-time.
- The skill parses the JSON response, ranking results based on a relevance score provided by the engine.
- It returns structured data including titles, URLs, and snippets for the agent to process within the Openclaw Skills framework.
SearXNG Search Skill Setup
To use this with Openclaw Skills, first set up SearXNG via Docker:
mkdir -p ~/Projects/searxng/searxng
cd ~/Projects/searxng
# Create your docker-compose.yml and settings.yml then start
docker compose up -d
Configure the required environment variable to point to your instance:
export SEARXNG_URL="http://localhost:8080"
Ensure that the settings.yml file has the JSON format enabled to allow Openclaw Skills to communicate with the API effectively.
SearXNG Search Skill Data Schema & Taxonomy
The skill returns highly structured JSON data to ensure compatibility with other Openclaw Skills components:
| Property | Description |
|---|---|
title |
The headline of the search result. |
url |
The direct link to the source page. |
content |
A snippet or description of the result content. |
engines |
Array of search engines that returned this specific result. |
score |
Relevance weighting (higher values indicate better matches). |
category |
The classification of the result (e.g., IT, News, Science). |
SearXNG Search Skill Advanced Features
- Support for category-specific filtering (IT, Science, News, etc.) for more targeted Openclaw Skills outputs.
- Multi-language and regional search support via localized query parameters.
- Seamless pagination for deep-dive research tasks using the pageno parameter.
- Custom shell function integration for rapid command-line searching.
- No external API keys or rate limits required since the infrastructure is entirely self-hosted.
- Aggregated relevance scoring from over 70 global search engines.
SKILL.md
Loading
Related Openclaw Skills
SearXNG Search Skill
A self-hosted, privacy-respecting metasearch skill that aggregates results from over 70 sources into a structured JSON API.
SearXNG Search
A privacy-respecting metasearch skill that aggregates results from over 70 engines into a single, self-hosted API for secure web research.
MoltGov
MoltGov is a democratic infrastructure layer for Moltbook AI agents to self-organize through citizenship, trust networks, and formal elections.
WeChat Send
A specialized automation tool for sending text messages through the native macOS WeChat desktop application using UI scripting.
LiquidSkills: Hyperliquid AI Agent Framework
LiquidSkills provides a specialized knowledge layer for AI agents to bridge the gap between stale training data and production Hyperliquid infrastructure.
ClawFlows
A command-line engine designed to search, install, and orchestrate multi-step automations by chaining various Openclaw Skills together.
Featured*
