How to Connect SAP Signavio to ChatGPT via MCP

💡 No install? Use cloud.anythingmcp.com directly. Sign in, click Connectors → SAP Signavio Process Manager, paste your credentials, mint an MCP API key and you're done. No Docker, no git clone, no local server.

SAP Signavio + ChatGPT

Read the SAP Signavio Process Manager workspace: folders, process models and revisions, exports as Signavio JSON or BPMN 2.0 XML, full-text search and the process dictionary. Session login with a technical user.

Prerequisites

See the full setup instructions baked into the connector (visible in the in-app store when you select the connector). The required environment variables for this connector are:

SIGNAVIO_URL, SIGNAVIO_EMAIL, SIGNAVIO_PASSWORD, SIGNAVIO_WORKSPACE_ID

Step 1 — Get credentials

1. Create a technical user in your SAP Signavio workspace that logs in with an email address and an SAP Signavio password, and give it the free API Edition licence. Such a user logs in with its SAP Signavio password even when the workspace uses single sign-on, but cannot use the web interface. A user with any other licence consumes a paid seat. If the workspace has no API Edition licences, ask SAP Support for them.
2. Give the user read access to the folders the AI should see. It only sees what its folder permissions allow.
3. Set SIGNAVIO_URL to the host you open Process Manager on, with https:// and without a path: EU https://editor.signavio.com, US https://app-us.signavio.com, Australia https://app-au.signavio.com, Canada https://app-ca.signavio.com, Japan https://app-jp.signavio.com, Singapore https://app-sgp.signavio.com, South Korea https://app-kr.signavio.com.
4. Put the user's email address in SIGNAVIO_EMAIL and its SAP Signavio password in SIGNAVIO_PASSWORD (not the single sign-on password of your company account).
5. In Process Manager open the Help menu → Workspace information and copy the Workspace ID (32 characters) into SIGNAVIO_WORKSPACE_ID. It is required: SAP Signavio refuses a login without it.
6. Install the connector and run sap_signavio_list_root_directories. It lists the root folders of the workspace (for example Shared documents and My documents), which shows the login works.

Sign-in: the Process Manager API has no API keys. The connector logs in as the technical user, keeps the session for its calls and logs in again when the session ends.

Read-only: this first version only reads. The AI browses folders, reads process models with their metadata and revisions, exports them and searches the workspace, but it cannot create, change, move or delete models, folders or dictionary entries.

What it reads: folders, process models (BPMN, EPC, value chains and more) with their revisions, full-text search across models, folders, shapes, files and comments, and the dictionary (organizational units, documents, activities, events and IT systems) with its categories. The AI starts with sap_signavio_list_root_directories and walks folders with sap_signavio_get_directory, or finds a model by name or content with sap_signavio_search.

Exports: sap_signavio_export_model_json returns the latest revision of a model as Signavio JSON and sap_signavio_export_model_bpmn as BPMN 2.0 XML. sap_signavio_get_revision_json and sap_signavio_get_revision_bpmn do the same for an older revision from sap_signavio_list_model_revisions. Large diagrams produce big exports, so the AI reads sap_signavio_get_model_info first and exports only the models it needs.

Limits: SAP allows 50 requests per 60 seconds and 5 parallel reads per workspace, so large folder trees are walked step by step. Search returns at most 250 results per call, the dictionary at most 1,000 entries.

Common errors: 401 means the login failed (check the email, password and workspace ID) or the session ended twice in a row. 403 means the user has no access to that folder.

Process Intelligence and Process Governance use separate APIs and are not covered.

Step 2 — Install the adapter

curl -fsSL https://raw.githubusercontent.com/HelpCode-ai/anythingmcp/main/docker-compose.quickstart.yml -o docker-compose.yml
printf 'JWT_SECRET=%s\nENCRYPTION_KEY=%s\n' "$(openssl rand -hex 32)" "$(openssl rand -hex 32)" > .env
docker compose up -d

Step 3 — Add the connector in ChatGPT

Your server URL: in AnythingMCP, open MCP Servers → the server this connector is on, and copy its URL (https://cloud.anythingmcp.com/mcp/…). Use it wherever this guide shows YOUR_SERVER_ID.

ChatGPT adds MCP servers as plugins on its Plugins page, on the web and on paid plans. Developer mode is no longer needed. In a Business workspace only admins and owners can add one; in Enterprise and Edu an admin can grant the Create plugins with MCPs permission.

1. Go to chatgpt.com/plugins, click +, then Add custom MCP server.
2. Enter a name, e.g. Name: SAP Signavio.
3. Under Connection, enter the Server URL https://cloud.anythingmcp.com/mcp/YOUR_SERVER_ID and keep Authentication on OAuth.
4. Select I understand and want to continue, then Create as a plugin. When ChatGPT asks, sign in to AnythingMCP and allow access.
5. Install the new plugin from your personal plugins, then type @ in a chat and pick it.

Available tools

Example prompts

• "Find every process model that mentions credit check and tell me which folder each one is in."
• "Export our order-to-cash process as BPMN 2.0 XML and list its tasks, events and gateways."
• "Compare the latest revision of the purchase requisition process with the previous one and tell me what changed."
• "List the IT systems in our process dictionary with their descriptions."

FAQ

Does the free ChatGPT plan work with custom MCP servers? OpenAI documents custom MCP servers for paid plans only; the Free plan is not on its list.

Next steps

• Connector landing page
• Connect to Claude
• Connect to Cursor
• Connect to Gemini
• Connect to Meta Muse