ask-search for Openclaw

A self-hosted, privacy-centric web search tool that aggregates results from major engines like Google and Bing without requiring API keys.

sexychina
v1.0.0
Mar 14, 2026
0
1.1k
0

Install & Download

1. ClawHub CLI

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

npx clawhub@latest install ask-search

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 ask-search 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 ask-search?

ask-search is a powerful search utility designed for the Openclaw Skills ecosystem that leverages a self-hosted SearxNG instance. By aggregating results from Google, Bing, DuckDuckGo, and Brave, it provides AI agents with a broad window into the live web while maintaining total privacy.

This skill is particularly valuable for developers who want to avoid the cost and tracking associated with proprietary search APIs. It delivers structured data including titles, snippets, and URLs, allowing for seamless integration into automated research and data retrieval pipelines.

ask-search Use Cases

  • Performing real-time market research and competitive analysis without tracking.
  • Gathering the latest news or scientific papers using specific category filters.
  • Extracting raw URL lists for downstream processing by web scraping tools.
  • Enabling AI agents to answer current-events questions with zero API overhead.

How ask-search Works

  1. The user or agent initiates a search command with specific keywords and optional filters like language or category.
  2. The skill queries a local or remote SearxNG instance, which concurrently fetches data from multiple search engines.
  3. Results are aggregated, cleaned, and returned as a structured list of snippets or raw JSON.
  4. The AI agent evaluates the snippets to decide if it has enough information or if it needs to perform a full page fetch via a secondary tool.

ask-search Setup

To use this skill within the Openclaw Skills framework, ensure you have a SearxNG instance running. By default, it looks for the service on port 8080.

# Install the skill (ensure SearxNG is running locally)
# If your instance is on a custom port, set the environment variable:
export SEARXNG_URL="http://localhost:8888"

# Test the search
ask-search "Openclaw Skills documentation"

ask-search Data Schema & Taxonomy

The skill returns structured search data that can be consumed as plain text or JSON.

Attribute Description
Title The clickable headline of the search result.
URL The direct link to the source content.
Snippet A brief summary or excerpt from the page.
Metadata Engine source and category information (available in JSON mode).

ask-search Advanced Features

  • Multi-engine aggregation: Combine results from Google, Bing, DuckDuckGo, and Brave simultaneously.
  • Category scoping: Restrict searches to news, science, images, or general web content.
  • Language localization: Use the --lang flag to retrieve region-specific results (e.g., zh-CN, en, ja).
  • Headless integration: Use the --urls-only flag to pipe search results directly into web fetching tools for automated scraping.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*