Make Your AI a Search Engine for Your Team's Internal Docs

Servers You'll Need

Step-by-Step Setup

1. 1

   Choose your knowledge source

   If your docs live in Notion, use the Notion MCP server. Google Drive → Google Drive server. Local Markdown files → filesystem server. You can use multiple sources simultaneously.

2. 2

   Set up API credentials

   For Notion: create an integration at notion.so/my-integrations and connect it to the pages you want searchable. For Google Drive: create OAuth2 credentials in Google Cloud Console.

3. 3

   Install and configure the server

   Install the relevant package and add it to your config with the appropriate API keys or OAuth tokens.

4. 4

   Test with real questions

   Ask your AI questions you'd normally look up in your docs. It will search the connected source and return answers with source references.

Example Prompts to Try

• “How do we handle customer refunds according to our policy docs?”
• “Find all onboarding documents in our Notion workspace”
• “What's the deployment process for the frontend? Search our runbooks”
• “Summarize our Q4 roadmap document”
• “What did we decide about the auth redesign in last week's meeting notes?”

Tips

• For Notion, only share the specific pages your AI needs — don't grant access to your entire workspace.
• Use the Memory MCP server alongside these to let the AI remember key facts across sessions.
• Tag or label important docs in Notion/Drive to make them easier for the AI to find.

FAQ

Related Use Cases