spring-boot-ai-confluence

This MCP server provides a set of Confluence-related tools that are exposed to MCP clients via the Spring Boot AI integration. The server connects to Confluence Cloud and registers callable operations such as listing Confluence spaces, retrieving pages, and creating or updating documents. Tools are annotated with @Tool and are registered through ToolCallbackProvider so they appear in MCP clients as available actions. To use, run the server and connect with any MCP client (e.g., Claude desktop app). The client can then invoke the registered tools by name and pass the required parameters, and the server will return structured results (e.g., lists of spaces, page summaries, or new document IDs).

Typical workflow:

• Start the server with the required Confluence credentials.
• In your MCP client, request the list of available tools to see supported operations (e.g., listSpaces, createDocument).
• Call a tool with the appropriate inputs (such as a space key, page title, or content body).
• The server processes the request against Confluence Cloud and returns results or created resource identifiers for follow-up actions.

Prerequisites:

• Java (JDK 17+) installed on your machine
• Maven or Gradle for building the Spring Boot project
• Confluence Cloud account with API access and a valid API token
• Access to the repository containing the Confluence MCP Server example

Installation steps:

1. Clone the repository: git clone https://github.com/your-org/bootcamptoprod-spring-boot-ai-confluence-mcp-server.git
2. Navigate to the project directory: cd bootcamptoprod-spring-boot-ai-confluence-mcp-server
3. Build the project (Maven): mvn clean package -DskipTests // or using Gradle if configured: ./gradlew build
4. Ensure you have Confluence credentials and set them in environment variables or application properties as described in the README (CONFLUENCE_BASE_URL, CONFLUENCE_API_TOKEN, CONFLUENCE_USERNAME).
5. Run the server: java -jar target/confluence-mcp-server.jar
6. Verify the server starts and is reachable by your MCP client, then load available tools and begin issuing commands.

Tips and caveats:

• Ensure your Confluence Cloud API token has the necessary permissions for the operations you intend to expose (listing spaces, creating/updating pages, etc.).
• Store sensitive credentials securely; avoid committing tokens in code or configuration files.
• If you see authentication errors, double-check the CONFLUENCE_BASE_URL and API token scopes.
• The server exposes tools via the @Tool annotation and registers them with ToolCallbackProvider; add or modify tools in the codebase to customize the MCP surface.
• For testing, you can use any MCP client to explore tool names, input schemas, and response formats.
• Logs are your friend: monitor stdout/stderr for hints about tool registration or Confluence API failures.