Create an MCP (Model Context Protocol) server to access dotCMS APIs

Parent Issue

No response

User Story

As a content or development team member, I want to be able to ask an LLM to perform tasks within dotCMS, so I can analyze my content and make changes to it more easily.

Anthropic has created the Model Context Protocol, which enables Anthropic AI tools (like Claude) to access data from and perform actions on third party systems. We need to understand the work required to create and stand up an MCP so Claude can read and write dotCMS content, and read Content Analytics data.

The first pass should identify the work involved to support the following actions in the MCP server (via the appropriate APIs):

• Query Content Analytics data (via the Content Analytics API)
• Read dotCMS content (via GraphQL)
• Execute Workflow Actions (via the Workflows API)
  • Note: This also requires being able to identify available Workflow Actions (via the Workflows API), so Claude can know what actions it can take on a given piece of content

Acceptance Criteria

New cards are created, with full labels and estimates, for each of the following:

• Implement the base MCP
• Support MCP Content Analytics queries
• Support MCP Content reads
• Support MCP Workflow execution

Proposed Objective

User Experience

Proposed Priority

Priority 2 - Important

External Links... Slack Conversations, Support Tickets, Figma Designs, etc.

No response

Assumptions & Initiation Needs

No response

Quality Assurance Notes & Workarounds

No response

Sub-Tasks & Estimates

No response

Start by mapping the Model Context Protocol work to the Content Analytics API, GraphQL content reads, and the Workflows API, including discovery of available Workflow Actions. Define the scope and estimates for the base MCP and each API integration. Done means four new cards exist with complete labels and estimates, as specified in the acceptance criteria.