Universal Search for Openclaw

A high-quality search API for Openclaw Skills providing structured results with credibility scoring and cross-validation.

cattei
v1.0.0
Mar 4, 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 universal-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 universal-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 Universal Search?

Universal Search is a specialized search interface developed by Sun Yongle, designed specifically for AI assistants and coding agents. It provides more than just raw links; it delivers high-quality, structured information including summaries, detailed explanations with source annotations, and cross-validation scores to ensure accuracy.

By integrating this into Openclaw Skills, developers can empower their agents with the ability to research complex topics with a high degree of confidence and factual reliability. The skill is optimized for AI consumption, ensuring that the data returned is easy for Large Language Models to parse and utilize in decision-making processes.

Universal Search Use Cases

  • Fact-checking and cross-referencing information during agentic workflows.
  • Gathering detailed summaries and source-backed explanations for technical queries.
  • Automating research tasks that require a credibility assessment of web sources.
  • Integrating high-fidelity search capabilities into AI-driven coding assistants for real-time documentation lookup.

How Universal Search Works

  1. The user or agent triggers the search script via the CLI with a specific query.
  2. The skill interfaces with the Universal Search API using the configured endpoint and authorization token.
  3. The API performs a comprehensive web search, aggregating and synthesizing data from multiple high-quality sources.
  4. Results are processed for credibility, scoring them on a scale of 0 to 10 based on cross-validation.
  5. A structured output is returned to the agent, containing the answer summary, citations, and verification scores.

Universal Search Setup

To get started with this component of Openclaw Skills, ensure you have Python 3 installed. Set the required environment variables in your shell or configuration file:

export UNIVERSAL_SEARCH_URL="https://49srjp57sf.coze.site/run"
export UNIVERSAL_SEARCH_TOKEN="your_bearer_token"

You can then perform a search using the following command:

python3 scripts/search.py "your search query" --json

Universal Search Data Schema & Taxonomy

The skill organizes its output into a highly structured format to support automated parsing. The data includes:

Attribute Description
Answer Summary A concise, high-level overview of the findings.
Detailed Explanation A comprehensive breakdown with specific source markers.
Source List A full list of reference URLs for transparency.
Verification Score A credibility rating from 0 (unreliable) to 10 (highly credible).

Universal Search Advanced Features

  • Configurable timeouts (up to 120 seconds) for deep-crawl research tasks.
  • JSON output mode for seamless integration into multi-agent pipelines and data processing workflows.
  • Built-in cross-validation logic to ensure facts are supported by multiple independent sources.
  • Specifically tuned return formats for AI assistant context windows.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*