WebSearch (SearXNG) for Openclaw

A privacy-first search skill that allows AI agents to query the web through a local SearXNG instance without relying on paid APIs.

netobasilio
v1.0.0
Feb 28, 2026
0
3.8k
0

Install & Download

1. ClawHub CLI

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

npx clawhub@latest install websearch

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 websearch 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 WebSearch (SearXNG)?

WebSearch is a powerful integration for Openclaw Skills that enables your AI agent to access real-time web information via a self-hosted SearXNG instance. By utilizing a local search engine, developers can provide their agents with up-to-date context and research capabilities while maintaining data sovereignty and avoiding the costs associated with commercial search APIs.

This skill is built for developers who require high-performance search capabilities within their agent workflows. It bridges the gap between static model knowledge and the ever-changing web, allowing for more accurate and context-aware responses in a secure, sandboxed environment.

WebSearch (SearXNG) Use Cases

  • Researching the latest software releases and technical documentation.
  • Gathering real-time data for market analysis or technical comparisons.
  • Providing agents with current event context to improve decision-making.
  • Reducing operational costs by replacing paid search providers with this Openclaw Skills integration.

How WebSearch (SearXNG) Works

  1. The AI agent receives a prompt requiring external information and triggers the WebSearch skill.
  2. The skill calls the local /usr/local/bin/websearch command on the host machine.
  3. The command forwards the query to a pre-configured, running SearXNG instance.
  4. SearXNG aggregates results from various web sources and returns the data.
  5. The skill captures the results and presents them to the agent for processing.

WebSearch (SearXNG) Setup

To integrate this search capability into your Openclaw Skills environment, ensure you have a functional SearXNG instance. Then, set up the execution binary:

# Ensure the websearch script is in the correct path
sudo cp path/to/websearch /usr/local/bin/websearch

# Make the script executable
sudo chmod +x /usr/local/bin/websearch

Ensure that the script is configured to point to your specific SearXNG URL.

WebSearch (SearXNG) Data Schema & Taxonomy

The skill utilizes a simple input-output schema to facilitate fast information retrieval:

Attribute Type Description
query string The keyword or natural language string to search for.
results string The raw data returned from SearXNG, typically in JSON or text format.

WebSearch (SearXNG) Advanced Features

  • Sandboxed execution ensures that search queries do not compromise host security.
  • Low-latency local communication with self-hosted search infrastructure.
  • Flexible output handling allows for raw JSON parsing or plain text summaries.
  • Can be combined with other Openclaw Skills for complex multi-agent research pipelines.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*