BookStack MCP Server
Provides comprehensive tools for managing BookStack instances, including full CRUD operations for books, chapters, and pages. It features advanced image gallery management with URL upload support and full-text search capabilities across all content entities.
README
BookStack MCP Server
This repository hosts a Python FastMCP-based server that exposes consolidated tools for managing a BookStack instance. The flagship capabilities are the image gallery management workflows that power authoring experiences in downstream MCP clients.
⚠️ DEPRECATION NOTICE: The TypeScript/mcp-framework server (
src/directory) is deprecated and no longer maintained. All development has moved to the Python FastMCP server (fastmcp_server/directory). Please migrate to the Python server for the latest features and bug fixes.
Quick start
# Install Python dependencies for the FastMCP server
pip install -r fastmcp_server/requirements.txt
Launch the FastMCP server after exporting your BookStack credentials (see below):
cd fastmcp_server
python3 -m fastmcp_server
Required environment
Copy .env.example to .env and populate these variables before invoking any BookStack tools:
BS_URL=https://your-bookstack.example.com
BS_TOKEN_ID=...
BS_TOKEN_SECRET=...
The API token must belong to a user that can view and manage the image gallery. Local helper scripts use set -a && source .env so the values apply to ad-hoc Python snippets as well.
BookStack tools
The Python FastMCP server provides comprehensive BookStack management through consolidated tools:
Content Management
bookstack_content_crud— unified CRUD operations for books, bookshelves, chapters, and pages (Letta-compatible)bookstack_list_content— list and filter content entities with paginationbookstack_search— full-text search across BookStack contentbookstack_batch_operations— bulk create, update, and delete operations
Image Gallery Management
bookstack_manage_images— unified create/read/update/delete/list interface for imagesbookstack_search_images— advanced discovery with extension, date, size, and usage filters
All tools are registered by fastmcp_server/bookstack/tools.py and surfaced automatically when the FastMCP server starts.
📘 Letta Compatibility: If you're using Letta as your MCP client, please read docs/LETTA_COMPATIBILITY.md for important compatibility requirements and best practices.
Image uploads from URLs
bookstack_manage_images accepts three input shapes for the image/new_image fields during create and update operations:
- Plain base64 strings
- Data URLs (
data:image/png;base64,...) - HTTP or HTTPS URLs
When a URL is supplied the tool:
- Streams the remote image with a 30 second timeout and a 50 MB limit
- Restricts schemes to HTTP/HTTPS to avoid SSRF
- Validates the MIME type against BookStack's accepted formats (jpeg, png, gif, webp, bmp, tiff, svg+xml)
- Infers a filename from the URL path when one is not supplied
Required BookStack parameters
BookStack's POST /api/image-gallery endpoint enforces two additional fields beyond the binary payload:
type— must begalleryfor standard content images (usedrawioonly when uploading diagrams.net PNGs)uploaded_to— the numeric page ID to attach the image to. BookStack rejects uploads without a real page context.
The tool surfaces these as optional inputs named image_type and uploaded_to. Default values of gallery and 0 preserve backward compatibility while allowing callers to target specific pages when required.
Manual verification against a live instance
After exporting your environment variables you can confirm an end-to-end URL upload with the following snippet (replace PAGE_ID with an existing page id):
cd /opt/stacks/bookstack-mcp/Bookstack-MCP
set -a && source .env && set +a
python3 - <<'PY'
import asyncio, json, time
from fastmcp import FastMCP
from fastmcp_server.bookstack.tools import register_bookstack_tools
TEST_IMAGE_URL = "https://upload.wikimedia.org/wikipedia/commons/4/47/PNG_transparency_demonstration_1.png"
PAGE_ID = 39 # replace with a page id from your BookStack instance
async def main():
mcp = FastMCP("manual-test")
register_bookstack_tools(mcp)
tool = await mcp.get_tool("bookstack_manage_images")
result = await tool.run({
"operation": "create",
"name": f"URL Upload Test {int(time.time())}",
"image": TEST_IMAGE_URL,
"uploaded_to": PAGE_ID,
})
print(json.dumps(json.loads(result.content[0].text), indent=2))
asyncio.run(main())
PY
You should receive a JSON payload describing the uploaded image, including thumbnails and the uploaded_to identifier. A 422 error means BookStack rejected the request (common causes: missing uploaded_to, disallowed MIME type, image exceeding the 50 MB limit). A 404 response typically indicates the API token lacks gallery permissions.
Testing
Run the Python unit tests for the BookStack tools:
cd fastmcp_server
python3 -m pytest tests/test_manage_images.py -v
The suite covers URL handling, timeout and size enforcement, invalid scheme rejection, and the forwarding of type/uploaded_to metadata.
Additional references
- FastMCP docs: https://gofastmcp.com/
- BookStack API reference: https://www.bookstackapp.com/docs/api/
- Product requirements for the image gallery tools:
docs/PRD-Image-Gallery-Management.md
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.