Skip to main content
Watch: I Built an AI App Without Leaving Databricks

Content and CMS - official IBM server

IBM Content Services MCP server

MCP server that exposes IBM FileNet Content Manager capabilities to AI models in four configurations: core document and folder management, property extraction and classification, legal hold, and AI document insight. It runs locally; the notes below cover how it authenticates.

Last verified 2 sources

Install

Generic

JSON
{
  "mcpServers": {
    "core-cs-mcp-server": {
      "command": "uvx",
      "args": [
        "--from",
        "git+https://github.com/ibm-ecm/ibm-content-services-mcp-server",
        "core-cs-mcp-server"
      ],
      "env": {
        "USERNAME": "your_username",
        "PASSWORD": "your_password",
        "SERVER_URL": "https://your-graphql-server/content-services-graphql/graphql",
        "OBJECT_STORE": "your_object_store"
      }
    }
  }
}

Notes and caveats

  • README opens with 'Do not use in production'.
  • Connects to the Content Services GraphQL API with username/password and object store settings.

FAQ

What does the IBM Content Services MCP server do?

MCP server that exposes IBM FileNet Content Manager capabilities to AI models in four configurations: core document and folder management, property extraction and classification, legal hold, and AI document insight.

How do I add the IBM Content Services MCP server to an MCP client?

IBM documents a Generic config; copy it from the Generic section of this page.

Does the IBM Content Services MCP server need authentication?

It uses a scheme other than OAuth or an API key, such as the session of the app or instance it runs in; the notes on this page describe it. Transport: stdio. Status: Preview, last checked October 17, 2026.

Sources

Last verified against 2 IBM pages, then re-checked independently.

Newsletter

New MCP servers, the week they ship

One email per week. New official MCP servers, the coding tools that support them and how developers wire them up.