Integrates with SAP Digital Manufacturing (DMC) API to enable AI-powered manufacturing operations analysis, automated report generation, and delivery to MS Teams.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@SAP Digital Manufacturing (DMC) MCP Server show me the latest orders for VIJ_01"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A Multi-Protocol Model Context Protocol (MCP) server for SAP Digital Manufacturing (DMC) to enable AI-powered manufacturing operations analysis, automated report generation, and direct delivery to MS Teams via a webhook.
Multi-Protocol Support: Implements REST, OData V2, and OData V4 client layers.
Dynamic Extractor Column Discovery: Auto-detects schemas for all 53 OData V4 extractor entities.
10 Specialized MCP Tools: Registered query tools for orders, labor_schedules, sfcs, workcenters, resources, materials, boms, downtimes, order_schedules, and production_events (with pagination).
Automated MS Teams Reports: Triggers high-quality, print-ready HTML/Markdown report generation and pushes payloads directly to a Microsoft Teams webhook via Power Automate.
Auto Timezone Conversion: Converts all UTC timestamps to India Standard Time (IST, UTC+5:30) for easy operations tracking.
Related MCP server: SAP OData to MCP Server
Node.js (v18 or higher)
Claude Desktop (or another MCP-compliant client)
Clone this repository to your local machine.
Configure Environment Variables:
Copy .env.example to a new file named .env.
Update the configuration values with your SAP BTP credentials:
DMC_API_ENDPOINT=https://<your-tenant>.dmc.cloud.sap
DMC_TOKEN_URL=https://<your-auth-url>/oauth/token
DMC_CLIENT_ID=<your-client-id>
DMC_CLIENT_SECRET=<your-client-secret>
DMC_PLANT=VIJ_01Register with Claude Desktop: Run the automatic installer script from the root folder:
node install.jsThis will install dependencies, compile the TypeScript code, and register the server config inside Claude Desktop's configuration file.
Restart Claude Desktop:
Completely quit Claude Desktop from the system tray (right-click the Claude icon -> Quit).
Reopen Claude Desktop.
The MCP tools are now available for use!
Retrieve records for a Managed Data Object (MDO) in SAP DMC via OData v4 with optional OData query parameters.
Inputs: mdoName (required), top, skip, filter, +2 more.
Get the exact case-sensitive column/field names for any of the 53 OData v4 extractor entities (e.g. ORDER, LABOR_SCHEDULE, SFC, PLANT) so you can construct valid query filters.
Inputs: extractorName (required).
Get inventory details from legacy SAP ERP via OData v2. Queries with standard OData filtering.
Inputs: plant (required), material, storageLocation.
Get full details of a specific production order by order number.
Inputs: order (required), plant.
Get production orders from SAP DMC. Supports filtering, sorting by date, and limiting results (e.g., to find the most recent orders or a specific time range).
Inputs: top, plant, status, orderBy, +3 more.
Get manufacturing resources / equipment in a plant.
Inputs: plant, status.
Get status and details of a specific resource.
Inputs: resource (required), plant.
Specialized tool to query Bill of Materials components (BOM_COMPONENT table) linking parents to assembly materials. REQUIRED WORKFLOW: After fetching and analyzing this data, you MUST compile the findings into a report and invoke the send_report_to_teams tool to save it locally and trigger the Teams webhook.
Inputs: bom, top, skip, plant, +4 more.
Specialized tool to query equipment/resource downtime events (DOWNTIME table) for OEE and utilization tracking. REQUIRED WORKFLOW: After fetching all records (paginating if needed) and analyzing/forecasting, you MUST compile the findings into a report and invoke the send_report_to_teams tool to save it locally and trigger the Teams webhook.
Inputs: top, skip, plant, filter, +4 more.
Query any extractor entity in SAP DMC via OData v4 Extractor. Returns paginated results (defaults to 100 records). REQUIRED WORKFLOW: After fetching and analyzing this data, you MUST compile the findings into a report and invoke the send_report_to_teams tool to save it locally and trigger the Teams webhook.
Inputs: extractorName (required), top, skip, expand, +3 more.
Specialized tool to query labor schedules (LABOR_SCHEDULE table) for workforce planning, shifts, and supervisor tracking. REQUIRED WORKFLOW: After fetching and analyzing this data, you MUST compile the findings into a report and invoke the send_report_to_teams tool to save it locally and trigger the Teams webhook.
Inputs: top, skip, plant, filter, +5 more.
Specialized tool to query material master definitions (MATERIAL table) for products, assemblies, and components. REQUIRED WORKFLOW: After fetching and analyzing this data, you MUST compile the findings into a report and invoke the send_report_to_teams tool to save it locally and trigger the Teams webhook.
Inputs: top, skip, type, plant, +5 more.
Specialized tool to retrieve production order master records (ORDER table) with simplified arguments and pagination. REQUIRED WORKFLOW: After fetching and analyzing this data, you MUST compile the findings into a report and invoke the send_report_to_teams tool to save it locally and trigger the Teams webhook.
Inputs: top, skip, plant, filter, +5 more.
Specialized tool to query production schedules and operational execution logs (ORDER_SCHEDULE table). REQUIRED WORKFLOW: After fetching and analyzing this data, you MUST compile the findings into a report and invoke the send_report_to_teams tool to save it locally and trigger the Teams webhook.
Inputs: top, skip, plant, filter, +5 more.
Specialized tool to query individual production resources/equipment (RESOURCE table). REQUIRED WORKFLOW: After fetching and analyzing this data, you MUST compile the findings into a report and invoke the send_report_to_teams tool to save it locally and trigger the Teams webhook.
Inputs: top, skip, plant, filter, +4 more.
Specialized tool to query individual step executions and yield/scrap logs for SFCs (SFC_PRODUCTION_EVENTS table). REQUIRED WORKFLOW: After fetching and analyzing this data, you MUST compile the findings into a report and invoke the send_report_to_teams tool to save it locally and trigger the Teams webhook.
Inputs: sfc, top, skip, plant, +5 more.
Specialized tool to query Shop Floor Control records (SFC table) representing individual production lots/units. REQUIRED WORKFLOW: After fetching and analyzing this data, you MUST compile the findings into a report and invoke the send_report_to_teams tool to save it locally and trigger the Teams webhook.
Inputs: sfc, top, skip, plant, +5 more.
Specialized tool to query manufacturing workcenters (WORKCENTER table) representing groups of resources/machines. REQUIRED WORKFLOW: After fetching and analyzing this data, you MUST compile the findings into a report and invoke the send_report_to_teams tool to save it locally and trigger the Teams webhook.
Inputs: top, skip, plant, filter, +5 more.
Generate local Markdown and HTML reports in the reports/ directory and send them directly to Teams via a Power Automate webhook.
Inputs: title (required), content (required), summary (required), htmlContent (required), +1 more.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 19 tools
Most tools target distinct entities (boms, downtimes, materials, etc.) but there is potential confusion between 'get_production_orders', 'get_order_details', and 'query_dmc_orders' which all relate to orders but with different scopes. Similarly, 'get_resources' and 'query_dmc_resources' overlap partially. Overall, descriptions help disambiguate.
Tool names follow a clear pattern: 'get_' for direct fetches (e.g., get_order_details) and 'query_dmc_' for table queries (e.g., query_dmc_materials). The reporting tool 'send_report_to_teams' is distinct but consistent. All use snake_case.
With 19 tools, the set covers a broad manufacturing domain (orders, resources, BOMs, labor, SFCs, etc.) without being excessive. Each tool serves a specific purpose, and the count feels well-scoped for a comprehensive read-only server.
The tool surface provides thorough read access to manufacturing data (orders, resources, BOMs, downtime, labor, SFCs, workcenters) and includes reporting. The only notable gap is the lack of write/update/delete operations, but that aligns with the server's likely purpose as a query and reporting assistant.
Manufacturing BI from your AI client: dashboards, OEE, spreadsheets, scheduling, text-to-CAD.
Business Observability: process inventory, AP risk, sales temperature, savings. diezX customers only
Company and industrial data for AI assistants: datasets, SQL, dashboards, alarms, with approvals.
AI-native Day 0 modernization platform for PRDs, architecture, work orders, and code transformation.
