Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
- Added OpenClaw metadata (`openclaw.requires.env` and `openclaw.primaryEnv`) to SKILL.md for enhanced environment variable specification. - No changes to functionality or code; documentation updated only.
What This Skill Does
LangSearch provides a free API for web search and semantic reranking. It uses hybrid search combining keyword matching with vector-based similarity to retrieve current web documents. Built for RAG pipelines and LLM applications that need up-to-date real-world context.
Free tier combines hybrid keyword and vector search with built-in semantic reranking, removing the need to wire together separate search and reranking services.
When to use it
- Fetching recent news or events for an AI chatbot response
- Building a RAG pipeline that pulls live web content
- Reranking search results by semantic relevance to a specific query
- Answering questions about topics beyond an LLM's training cutoff
- Aggregating results from multiple queries into a research context
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: What are the best practices for deploying ML models in production?
- 1Sends query to the LangSearch web search endpoint with summary enabled
- 2Retrieves top 10 results including titles, snippets, and markdown summaries
- 3Prepares result snippets as documents and passes them to the rerank endpoint
- 4Returns top 5 semantically reranked documents as grounded context
- 5Formats context for the LLM to generate a current, sourced response
Top 5 reranked web results formatted as context for LLM response generation
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.