datasphere-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| DSP_ENV_FILE | No | Path to the .env file. Required if the MCP client does not set the working directory correctly. | |
| DSP_CLI_ENTRY | No | Path to the SAP Datasphere CLI entry point (terminal.js). Required if the MCP client does not set the working directory correctly. | |
| DSP_MOCK_MODE | No | Set to 'true' to run the server in mock mode using local fixtures instead of connecting to SAP Datasphere. Default is 'false'. | false |
| DSP_DEV_BASE_URL | No | The origin URL of the SAP Datasphere DEV tenant, e.g., https://YOUR-DEV-TENANT. | |
| DSP_DEV_AUTH_MODE | No | Authentication mode for the DEV tenant. Use 'refresh_token' for user-based OAuth (interactive usage) or 'client_credentials' for a technical user. | |
| DSP_DEV_CLIENT_ID | No | The OAuth client ID for the DEV tenant. | |
| DSP_DEV_TOKEN_URL | No | The OAuth token URL for the DEV tenant. | |
| DSP_DEV_SECRETS_FILE | No | Path to a local secrets JSON file exported from the SAP Datasphere CLI. Alternative to providing OAuth client credentials directly. | |
| DSP_DEV_CLIENT_SECRET | No | The OAuth client secret for the DEV tenant. | |
| DSP_DEV_REFRESH_TOKEN | No | The user refresh token for OAuth when using refresh_token authentication. | |
| DSP_DEV_SPACES_BACKEND | No | Backend to use for listing spaces when in non-mock mode. Valid values are 'cli' or 'catalog'. Default is 'cli'. | cli |
| DSP_RUN_DEV_INTEGRATION | No | Set to 'true' to enable integration tests against the real DEV tenant. Default is 'false'. | false |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| logging | {} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_spacesB | List accessible spaces. Explicit environment; bounded pagination. |
| list_objectsC | List modeling objects of a verified CLI type; follow next_offset for additional pages. |
| get_objectB | Read the original SAP CSN/JSON definition, preserving metadata except secret redaction. |
| get_dependenciesB | Infer bounded upstream CSN dependencies with evidence. DOWNSTREAM/BOTH are unsupported. complete=false means this is partial metadata extraction, not authoritative tenant lineage. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 4 tools
Each tool targets a distinct resource and operation: listing spaces, listing objects, reading one object, and inferring dependencies. Overlap between list_objects and get_object is minimal (collection vs single resource), and get_dependencies is clearly separate.
All names follow a consistent snake_case verb_noun pattern (list_spaces, list_objects, get_object, get_dependencies), with no deviations or mixed conventions.
Four tools are well-scoped for a read-only metadata and lineage connector; each covers a core operation without redundancy. The count sits comfortably within the typical 3-15 range.
Core read operations are present, but the surface lacks space-level detail retrieval, search, and any downstream dependency support (explicitly unsupported). The note that output is non-authoritative also signals partial coverage of typical lineage workflows.