SAP AI Core Documentation

You run a dedicated MCP server that provides semantic search and fast access to SAP AI Core documentation. It enables AI assistants to locate, retrieve, and understand SAP AI Core materials efficiently, improving accuracy and speed when answering questions or providing guidance.

How to use

You interact with the SAP AI Core Docs MCP Server through an MCP client. Start the local server, then point your client at the configured MCP server to perform semantic search, fetch full documentation pages, or retrieve topic-specific content. Use search to filter by category, request complete documents by path, or pull comprehensive topic overviews for quick reference.

How to install

Prerequisites you need before installing: Node.js 20.0.0 or higher and npm or yarn.

Step 1: Install dependencies and prepare the environment.

sudo apt-get update
sudo apt-get install -y nodejs npm
node -v
npm -v

Step 2: Set up the project and build the MCP server.

git clone <repository-url>
cd dlwr-dnl-ai-core-documentation-mcp
npm install

Step 3: Prepare the SAP AI Core docs as a submodule.

git submodule add https://github.com/SAP-docs/sap-artificial-intelligence.git docs/sap-artificial-intelligence
git submodule update --init --recursive

Step 4: Build the server so you can run it.

npm run build

Configuration

Claude Desktop configuration allows you to register the MCP server so your Claude instance can communicate with it.

{
  "mcpServers": {
    "sap-ai-core-docs": {
      "command": "node",
      "args": [
        "/absolute/path/to/dlwr-dnl-ai-core-documentation-mcp/build/index.js"
      ]
    }
  }
}

If you want to use a custom documentation location, specify an environment variable to point to the new docs folder.

{
  "mcpServers": {
    "sap-ai-core-docs": {
      "command": "node",
      "args": [
        "/absolute/path/to/dlwr-dnl-ai-core-documentation-mcp/build/index.js"
      ],
      "env": {
        "SAP_AI_CORE_DOCS_PATH": "/path/to/custom/docs"
      }
    }
  }
}

Notes on usage

The server indexes all markdown files from the SAP AI Core documentation repository at startup, enabling fast in-memory search with multi-field scoring and context extraction. It supports category filtering to narrow results to administration, development, integration, or concepts areas.

Troubleshooting

If you encounter startup or search performance issues, verify that dependencies are installed, the submodule is initialized, and the build completed successfully. Ensure your MCP client is configured to reach the local Node process and that the environment variable for a custom docs path is correctly set if you changed the default location.

Available tools

search_ai_core_docs

Semantically search SAP AI Core documentation with optional category and limit filters.

get_ai_core_document

Retrieve complete content of a specific documentation page by path.

get_ai_core_topic

Get comprehensive documentation for a specific SAP AI Core topic.

list_ai_core_categories

List all available categories and top documents.