hexagon-nodesBluebricks Docs for AI Agents

Give AI agents access to Bluebricks documentation via llms.txt for static context or the GitBook Docs MCP server for interactive, real-time search.

Bluebricks offers two ways to give AI agents access to the documentation: llms.txt for a lightweight, static overview you can feed into any LLM, and the GitBook Docs MCP server for interactive, real-time search and retrieval.

What is an llms.txt File?

Think of llms.txt as the robots.txt for the AI era. Instead of telling a crawler what not to index, it tells an LLM what it should prioritize.

Key Features

  • Markdown Format: Human-readable and machine-parsable.

  • Structured Overview: Contains a high-level summary and a list of categorized links.

  • Context-First: Points to "full" versions of documentation (llms-full.txt) that are optimized for LLM context windows, stripped of ads and navigation menus.

How to Work with llms.txt

The file typically lives at the root of your project or website (e.g., yourproject.com/llms.txt).

Step #1: Downloading the File

Step #2: Using it in Development

When you point an AI agent to a repository or a URL, the agent checks for this file. It uses the links provided to "pull" only the relevant Markdown files into its memory, rather than trying to scrape a complex HTML website.


GitBook Docs MCP

For interactive, real-time access to the documentation, you can connect your AI agent to the Bluebricks Docs MCP server. Instead of loading a static file, the agent searches and fetches specific pages on demand, so it always works with the latest content.

Endpoint URL:

Setup

chevron-rightVS Codehashtag

Add the following to your .vscode/mcp.json:

chevron-rightClaude Desktophashtag

Add the Bluebricks Docs MCP server to your Claude Desktop configuration:

Configuration file location:

  • ~/.claude/settings.json

chevron-rightCursorhashtag

Enable MCP support in Cursor settings:

  1. Open Cursor Settings (Cmd+, or Ctrl+,)

  2. Navigate to Extensions → MCP

  3. Add new server configuration:

    • Name: bluebricks-docs

    • Type: sse

    • URL: https://bluebricks.co/docs/~gitbook/mcp

chevron-rightOther MCP Clientshashtag

For any MCP-compatible client, configure with:

  • Name: bluebricks-docs

  • Type: SSE

  • URL: https://bluebricks.co/docs/~gitbook/mcp

Consult your client's documentation for specific configuration steps.

What agents can do

Once connected, your AI agent can:

  • Search the docs by keyword or topic to find relevant pages

  • Fetch specific pages to pull full content into context

  • Stay current with always up-to-date content, no manual re-downloads needed


llms.txt vs GitBook Docs MCP

circle-info

Not sure which approach to use? Here's a quick comparison.

llms.txt

GitBook Docs MCP

Access model

Static file download

Interactive search and fetch

Setup

Zero-setup, just point to the URL

Requires MCP client configuration

Content freshness

Snapshot at time of download

Always up to date

Best for

Quick context dumps, one-off queries, agents without MCP support

Ongoing development, multi-turn conversations, context-efficient workflows

Context usage

Loads full file into context

Fetches only the pages you need

Last updated

Was this helpful?