This MCP server exposes CAST Imaging application intelligence to MCP clients, including application inventories, architecture graphs, transactions and data flows, database schema details, object relationships, and quality/security findings. It acts as a bridge to CAST Imaging-APIs and returns paginated data for exploration and analysis.
From the registry: MCP server exposing CAST Imaging software architecture insights to AI agents
$ curl -H "x-api-key: <your-imaging-api-key>" http://<your-control-panel-host>:8090/imaging/apis/rest/readyPlease install the `imaging-mcp-server` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Install Docker on Linux** — A Linux system with Docker installed and running is required for the documented server installation path.
- **Have CAST Imaging-APIs reachable** — The CAST Imaging-APIs service must be running and reachable from the MCP server host. The README suggests verifying readiness with the health endpoint using your Imaging API key. Run: `curl -H "x-api-key: <your-imaging-api-key>" http://<your-control-panel-host>:8090/imaging/apis/rest/ready`
- **Generate a CAST Imaging API key** — Create an API key from your CAST Imaging user profile. Clients use this key in HTTP headers when connecting to the MCP server.
- **Download and extract the CAST Imaging MCP Server installer** — The server is distributed as a proprietary installer package that must be downloaded and extracted before configuration and startup.
Canonical MCP server config (HTTP transport):
- url: `http://<your-mcp-server-host>:<mcp-server-port>/mcp/`
- headers:
- `x-api-key`: `<your-imaging-api-key>`
- `x-user-tenant`: `<your-imaging-tenant>`
Note: This server is exposed over HTTP after installing and running the proprietary Docker-based package with ./run.sh --install. Server-side setup requires configuring app.config and .env before startup. Required server configuration values include HOST_CONTROL_PANEL and SERVICE_HOST. Common server settings documented in app.config/.env are: HOST_CONTROL_PANEL (required, Control Panel host), SERVICE_HOST (required, MCP server host), PORT_CONTROL_PANEL (optional, default 8098), MCP_SERVER_PORT (optional, default 8282), IMAGING_PAGE_SIZE (optional), IMAGING_DISPLAY_PAGE_SIZE (optional), IMAGING_CODE (optional), DEBUG_MODE (optional), IMAGING_DOMAIN (optional, default default), CONTROL_PANEL_SSL_ENABLED (optional, default false), MCP_TOOL_SURFACE_PROFILE (optional, default full), MCP_INTENTS_HIDDEN_FUNCTIONS (optional), and SSL_CA_BUNDLE (optional, path like /app/certificates/<file>.pem when using private/self-signed CAs).
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.HOST_CONTROL_PANELrequiredHost/IP of the CAST Imaging Control Panel service registry used to discover Imaging servicesSERVICE_HOSTrequiredHost/IP of the machine running the MCP serverMCP_SERVER_PORTTCP port the MCP server listens onPORT_CONTROL_PANELControl Panel registry portIMAGING_DOMAINImaging domain/tenant to query in multi-tenant setupsIMAGING_CODEControls whether source code can be accessed through the MCP serverCONTROL_PANEL_SSL_ENABLEDWhether Control Panel registry endpoints use HTTPSMCP_TOOL_SURFACE_PROFILESelects the exposed MCP tool profile (full or intents)MCP_INTENTS_HIDDEN_FUNCTIONSComma-separated structural functions hidden when using intents modeSSL_CA_BUNDLEAI orchestration with hive-mind swarms, neural networks, and 87 MCP tools for enterprise dev.
Persistent codebase knowledge graph. Survives session restarts and context compaction.
A powerful toolkit for coding, providing semantic retrieval and editing capabilities.