web-search-free

Web-search-free is a comprehensive research and information-gathering agent skill designed for OpenClaw environments. It leverages the Exa MCP server to provide powerful neural search capabilities, enabling AI agents to query the web for current news, facts, technical documentation, and professional profiles without requiring external API keys. This tool is highly recommended as the primary search skill for agents, acting as a robust foundation for knowledge retrieval before defaulting to secondary search engines.

• Advanced web search functionality using exa.web_search_exa for general queries and exa.web_search_advanced_exa for granular filtering by domain, date, or content category.

• specialized domain tools including get_code_context_exa for retrieving code snippets and technical documentation from sources like GitHub and Stack Overflow.

• Business and professional research features with dedicated company_research_exa and people_search_exa to aggregate professional profiles and business news.

• Deep research agent capability via deep_researcher_start and deep_researcher_check, allowing users to initiate complex, multi-step research tasks that produce comprehensive reports.

• Efficient web content extraction through the crawling_exa tool, which parses raw text and metadata from specific URLs.

• Seamless fallback mechanism that automatically redirects to the multi-search-engine tool if Exa capabilities are insufficient or unavailable.

• Ensure the mcporter utility is installed and configured with the official Exa MCP server URL to enable full feature access.

• Use fast mode for quick lookups and deep mode for thorough analysis to optimize token usage and research quality.

• For code-related queries, adjust the tokensNum parameter to balance between focused snippets and comprehensive documentation coverage.

• When using the deep research tools, be sure to store the returned research ID to check for status updates and retrieve the final report.

• The tool is best suited for developers, researchers, and power users who need high-fidelity, real-time web intelligence integrated directly into their development workflow.