Skip to main content
Nimble provides real-time web data access through a search API that browses the live web rather than relying on pre-built indexes. Unlike traditional search APIs, Nimble uses headless browsers to navigate websites in real-time—handling JavaScript rendering, dynamic content, pagination, and complex multi-step navigation flows. Key technical capabilities:
  • Live web browsing: Real-time access to current web content, not cached indexes
  • JavaScript rendering: Handles modern SPAs, lazy loading, and client-side rendering
  • Flexible modes: Fast mode for SERP data or deep mode for full content extraction
  • AI summaries: Optional LLM-generated answers alongside raw search results
  • Production-ready: Built for scale with async support and comprehensive error handling
For more information, visit nimbleway.com or explore the official Nimble documentation.

​
Installation and setup

The Nimble integration exists in its own partner package. You can install it with: 
pip install -U langchain-nimble
To use the package, you’ll need to set the NIMBLE_API_KEY environment variable to your Nimble API key. You can obtain an API key by signing up at Nimble.

​
Tools and toolkits

NimbleSearchTool

Real-time web search for agents. Supports deep mode for full content extraction, domain/date filtering, and optional AI-generated summaries.

NimbleExtractTool

Extract rendered content from specific URLs. Handles JavaScript-heavy sites with configurable parsing formats.

​
Retrievers

NimbleSearchRetriever

Search retriever with fast/deep modes. Navigate dynamic sites and extract full page content for RAG applications.

NimbleExtractRetriever

Content extraction retriever for known URLs. Returns structured data in plain text, markdown, or HTML formats.
Edit this page on GitHub or file an issue.
Connect these docs to Claude, VSCode, and more via MCP for real-time answers.