SearXNG Search for Openclaw

A privacy-respecting metasearch skill that allows AI agents to search the web, images, and news via a local SearXNG instance.

liujiang817
v1.0.0
Mar 21, 2026
0
673
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install my-searxng

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install my-searxng 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?

SearXNG Search is a robust skill designed for Openclaw Skills that enables private, self-hosted web search capabilities. By leveraging a local SearXNG instance, this skill aggregates results from multiple engines like Google, Bing, and DuckDuckGo without external API dependencies or tracking. It is specifically engineered for AI workflows, providing clean, token-efficient text output that helps agents find real-time information, news, and technical documentation while maintaining strict data privacy.

This skill is ideal for developers who want to give their agents internet access without relying on expensive or privacy-invasive third-party search APIs. Since it uses a zero-dependency Python backend, it integrates seamlessly into existing environments with minimal overhead.

SearXNG Search Use Cases

  • Researching real-time news and current events directly within an AI conversation.
  • Fetching technical documentation from IT or science-specific search categories.
  • Finding images or media related to a user's query without external tracking.
  • Conducting multi-engine web searches while maintaining a zero-dependency, privacy-focused footprint.

How SearXNG Search Works

  1. The AI agent detects a search trigger such as "search the web" or "find information".
  2. The agent executes the provided Python script, passing the query and optional parameters like category or language.
  3. The script sends a request to the configured local SearXNG instance's JSON API.
  4. SearXNG aggregates results from various search engines and returns them to the script.
  5. The script formats the results into a clean, token-optimized table or text format for the LLM to process within Openclaw Skills.

SearXNG Search Setup

First, ensure you have a running SearXNG instance. Then, configure the instance URL in the searxng.ini file located at {baseDir}/scripts/searxng.ini.

[searxng]
url = http://your-searxng-instance.com:port

Ensure Python 3 is installed. To test the search manually via the CLI, use the following command:

python3 {baseDir}/scripts/searxng.py search "your query"

SearXNG Search Data Schema & Taxonomy

The skill manages its connection via a local configuration file and outputs data formatted for LLM consumption.

Component Details
Configuration searxng.ini storing the instance URL.
Input Parameters Query string, result count (-n), category, language, and time-range.
Default Output A standard text table optimized for low token usage.
JSON Output Available via --format json for raw data processing.

SearXNG Search Advanced Features

  • Support for specialized search categories including IT, Science, News, and Images.
  • Time-range filtering allowing results from the last day, week, month, or year.
  • Multi-language support to tailor results to specific regions (e.g., zh-CN, en).
  • Zero-dependency architecture requires no pip install, ensuring high portability.
  • AI-optimized output constraints that default to 10 results to preserve context window space in Openclaw Skills.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*