NLWeb
Main reference implementation for NLWeb, implemented in Python.
NLWeb is an MCP server implementation that enables conversational interfaces for websites by exposing natural language query capabilities to AI agents. Built in Python, it leverages Schema.org structured data already present on over 100 million websites to provide semantic understanding of web content. The server implements MCP's Model Context Protocol, allowing agents to ask natural language questions and receive JSON responses in Schema.org vocabulary. It supports multiple vector databases (Qdrant, Milvus, Snowflake, Postgres, Elasticsearch) and LLM providers (OpenAI, DeepSeek, Gemini, Anthropic), making it a flexible foundation for building AI-powered web interfaces.
Key Features
Use Cases
- 01Adding conversational query interfaces to e-commerce sites with product catalogs
- 02Enabling AI agents to search recipe databases using natural language
- 03Creating chatbot interfaces for tourism websites with attraction and review data
- 04Building enterprise data assistants that query CRM and project management systems
- 05Deploying natural language APIs for content-heavy websites with Schema.org markup
- 06Routing agent queries across multiple NLWeb-enabled sites via AgentFinder discovery
Related MCP Servers
View moremarkitdown
Python tool for converting files and office documents to Markdown.
firecrawl
The context API to search, scrape, and interact with the web at scale. 🔥
prompts.chat
f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.
langflow
Langflow is a powerful tool for building and deploying AI-powered agents and workflows.
NLWeb — FAQ
What is the NLWeb MCP server?+
NLWeb is a Python-based MCP server that enables AI agents to query websites using natural language. It leverages existing Schema.org structured data on web pages and returns JSON responses, acting as a conversational interface layer for web content.
How do I install the NLWeb MCP server?+
Clone the repository from GitHub, configure your LLM provider and vector database in the config YAML files, then add the server to your MCP client configuration pointing to the AskAgent module. See the 'Hello world on your laptop' guide for detailed setup instructions.
Which AI clients work with NLWeb?+
NLWeb works with any MCP-compatible client such as Claude Desktop. It implements the Model Context Protocol standard, making it compatible with chatbots and AI assistants that support MCP server connections.
Do I need API keys to use NLWeb?+
Yes, you'll need API keys for your chosen LLM provider (OpenAI, DeepSeek, Gemini, Anthropic, etc.) and access credentials for your selected vector database if using cloud-hosted options like Azure AI Search or Snowflake.
Is NLWeb free to use?+
Yes, NLWeb is open source under the MIT License and free to use. However, you'll incur costs for the LLM API calls and vector database services you choose to integrate with it.
What types of websites work best with NLWeb?+
Websites with existing Schema.org markup or structured data work best, including sites with product listings, recipes, events, reviews, and articles. The server extracts semantic meaning from this structured content to answer natural language queries accurately.
How do I install NLWeb?+
Open the source repository on GitHub and follow its README. NLWeb is a mcp server — MCP Agents Market links you directly to the official repo.
Is NLWeb free?+
NLWeb is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.