online-search

Query the ProSearch engine for real-time internet results with freshness filters.

Updated May 4, 2026
One-click install
npx skills add https://github.com/234194027-cpu/xianclaw --skill online-search-234194027-cpu
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: online-search
Source: https://github.com/234194027-cpu/xianclaw/tree/main/resources/openclaw/config/skills/online-search
Command: npx skills add https://github.com/234194027-cpu/xianclaw --skill online-search-234194027-cpu

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill helps users obtain up-to-date internet information without relying on the AI to invent sources, ensuring that the displayed result items (titles, links, sources) come directly from the search service.

Core Features & Use Cases

  • Real-time web search with formatted result items: Uses ProSearch to query the internet and returns a pre-rendered message containing the top results.
  • Time freshness filtering for “latest/today/recent” queries: Supports freshness modes such as 24h, 7d, 30d, and 1y to reduce stale results.
  • Optional constrained search modes: Supports count-based results, site/domain filtering, industry filtering (gov/news/acad), and VR-card style structured data (e.g., weather/price), while enforcing parameter mutual exclusivity.

Quick Start

Ask the AI to search the web with recency, for example: search for the latest AI news from the last 7 days and show the top results with clickable links.

Frequently Asked Questions about online-search

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I get real-time web search results without AI source hallucination?

Real-time web search without AI source hallucination is achieved by querying the ProSearch engine and displaying server-rendered, clickable top entries. This ensures returned titles, links, and sources come directly from the internet search service rather than being invented by the AI.

Can I filter internet search results by freshness for recent news and policy updates?

Internet search results can be filtered by freshness using modes like 24h, 7d, 30d, and 1y. This freshness filtering reduces stale results, making it ideal for querying recent news, policy updates, and time-sensitive lookup requests.

How do I constrain web results to specific domains or industries like government and academic sites?

Web results can be constrained using optional search modes that support site and domain filtering. You can also filter by industry categories such as government, news, and academic domains to narrow down the scope of your internet search results.

Does the search support structured data cards for specific queries like weather or prices?

The search supports VR-card style structured data for specific queries like weather and prices. This feature is available alongside other constrained search modes, including count-based results and industry filtering, while enforcing parameter mutual exclusivity.

What is the best way to look up time-sensitive facts online without relying on outdated information?

The best way to look up time-sensitive facts is using an internet search service with anti-hallucination guarantees and optional freshness constraints. It queries the web in real-time and outputs clickable top entries verbatim from the backend search engine.

When should I not use anti-hallucination web search for my queries?

You should not use anti-hallucination web search when you need the AI to synthesize or summarize information, as this approach outputs the returned message field verbatim as the primary result list without AI-generated summaries or rewritten content.