Skip to main content
Once the server is running, you can interact with it via the Web Interface, CLI, or your AI Assistant.

​
Web Interface

If you are running the Standalone Server (Docker or npx), the web interface is available at: http://localhost:6280

Add Documentation

Submit URLs or file paths for indexing

Monitor Jobs

Watch scraping and indexing progress in real-time

Manage Libraries

View and delete indexed documentation

Search

Manually test search queries to see what the AI will see

​
Launching Web UI for Embedded Server

If you are using the Embedded Server (running inside your AI tool), it does not expose a web interface by default. You can launch a temporary web UI that connects to the same database: 
OPENAI_API_KEY="your-key" npx @arabold/docs-mcp-server@latest web --port 6281
Open http://localhost:6281. Stop the process (Ctrl+C) when finished.
The web UI connects to the same database as your embedded server, so changes are reflected immediately.

​
CLI Usage

You can manage the server using command-line tools.
If you are using the Embedded Server, ensure you don’t run concurrent write operations (scraping) if the database is locked.
npx @arabold/docs-mcp-server@latest list
See the CLI Reference for complete command documentation.

​
Scraping Local Files

You can index documentation from your local filesystem using file:// URLs. This works in both the Web UI and CLI.

​
Requirements

Text-Based Files

HTML, Markdown, JS, TS, Python, Java, etc.

Binary Files Ignored

PDF, images are not processed via file:// URLs
Docker Users: You must mount the local directory into the container first.

​
Examples

1

Single File

file:///Users/me/docs/index.html
2

Directory

file:///Users/me/docs/my-library

​
Docker Example

If your docs are in /absolute/path/to/docs on your host:
1

Mount the volume

docker run --rm \
  -v /absolute/path/to/docs:/docs:ro \
  -v docs-mcp-data:/data \
  -v docs-mcp-config:/config \
  -p 6280:6280 \
  ghcr.io/arabold/docs-mcp-server:latest
2

Scrape using the container path

In the Web UI or CLI, use: file:///docs

​
AI Assistant Usage

Once connected, your AI assistant (Claude, Cline, etc.) will have access to tools like scrape_docs and search_docs.

​
Example Prompts

Prompt:
“Please scrape the React documentation from https://react.dev/reference/react for library ‘react’ version ‘18.x’”
The AI will use the scrape_docs tool to index the documentation.
Prompt:
“How does the useState hook work in React? Please check the documentation.”
The AI will use search_docs to find relevant documentation and provide an answer based on the indexed content.
Prompt:
“What versions of Next.js documentation do we have indexed?”
The AI will use list_libraries to show all indexed libraries and their versions.

​
Common Workflows

Add Framework Docs

Learn how to index documentation from various sources

Search Documentation

Master search queries and version matching

Configure Embeddings

Set up semantic search with embedding models

CLI Commands

Explore all available CLI commands

​
Next Steps

1

Enable Semantic Search

Configure embedding models for better search results
2

Index Your Documentation

Learn about different source types and how to index them
3

Optimize Search

Master search queries to get the most relevant results

Build docs developers (and LLMs) love

Get started for freeTalk to us