web-search-pro

What This Skill Does

Web Search Pro is an advanced, multi-engine search utility designed for the OpenClaw AI agent ecosystem. Unlike standard search tools that rely on a single source, this skill orchestrates requests across Tavily, Exa, Serper, and SerpAPI. It bridges the gap between basic queries and research-heavy tasks by providing granular control over how information is retrieved. It supports sophisticated filters such as domain inclusion/exclusion, precise date-range filtering, and deep-search modes that prioritize high-quality, long-form content. Furthermore, it includes a robust content extraction engine, allowing agents to pull clean, readable text directly from URLs using either Tavily Extract or Exa’s livecrawl capabilities.

Installation

To add this skill to your OpenClaw environment, ensure you have the necessary API keys for your preferred search engines, then run the following command in your terminal:

clawhub install openclaw/skills/skills/zjianru/web-search-pro

After installation, configure your environment variables (e.g., TAVILY_API_KEY, EXA_API_KEY, SERPER_API_KEY, or SERPAPI_API_KEY) to enable the respective search engines. The system will automatically detect these keys and assign search tasks based on the optimal engine for your specific query type.

Use Cases

• Technical Research: Filter results to high-authority developer documentation like GitHub or StackOverflow while ignoring SEO spam.
• Market Intelligence: Track recent news or market movements using specific date ranges or relative time windows (e.g., "last week").
• Deep Analysis: Use the --deep flag to trigger neural/AI-optimized search processes that synthesize complex information from multiple sources.
• Content Scraping: Extract full article text from diverse web sources for summarization or internal knowledge base updates.
• Localized Search: Use the SerpAPI integration to perform specific queries on platforms like Baidu for better results when researching topics specific to the Chinese market.

Example Prompts

1. "Perform a deep search on recent advancements in transformer architectures from the last 7 days, excluding medium.com articles."
2. "Search for documentation regarding the new OpenClaw API features, focusing only on github.com and official docs."
3. "Extract the full content from this URL [https://example.com/article] and summarize the key technical findings into a bulleted list."

Tips & Limitations

To get the most out of Web Search Pro, rely on the automatic engine selection whenever possible, as the internal logic is optimized to choose the most cost-effective and accurate provider for your task. If you are scraping large volumes of content, be mindful of the API usage limits imposed by the free tiers of your chosen services. Note that domain filtering efficacy may vary slightly between engines; if one engine fails to filter a domain correctly, try switching to a provider with stronger native filtering support like Tavily. For programmatic usage, always pipe the output using the --json flag to ensure compatibility with your agent's data-processing pipelines.