/datasphere-mcp-tools

Display available MCP tools for SAP Datasphere integration.

SAP Datasphere MCP Tools (45 total)

The SAP Datasphere MCP Server (@mariodefe/sap-datasphere-mcp) provides 45 tools for direct tenant interaction.

GitHub: https://github.com/MarioDeFelipe/sap-datasphere-mcp Package: npm install -g @mariodefe/sap-datasphere-mcp

---

Foundation Tools (5)

test_connection

Purpose: Verify connectivity to SAP Datasphere tenant Returns: Connection status, latency, tenant URL Use when: Setting up MCP integration, troubleshooting connectivity

get_current_user

Purpose: Get authenticated user details Returns: User ID, email, permissions, authentication status Use when: Verifying authentication, checking user context

get_tenant_info

Purpose: Retrieve tenant configuration and capabilities Returns: Tenant URL, version, features, region Use when: Discovering tenant features, checking version compatibility

get_available_scopes

Purpose: List available permission scopes Returns: Scopes array with descriptions Use when: Understanding permission model, configuring OAuth clients

list_spaces

Purpose: Discover all spaces in tenant Returns: Space names, IDs, storage info, owners Use when: Finding available spaces, discovering data domains

---

Catalog & Asset Tools (4)

search_catalog

Purpose: Search business catalog for objects Returns: Matching objects with metadata Use when: Finding existing views, tables, models by name or description

find_assets_by_column

Purpose: Find data assets by column name Returns: Objects containing specified column Use when: Discovering which tables/views contain specific columns

analyze_column_distribution

Purpose: Profile data distribution in columns Returns: Statistics, distinct values, nulls, distribution Use when: Data quality analysis, understanding data patterns

get_asset_metadata

Purpose: Retrieve detailed asset metadata Returns: Complete object definition, lineage, usage Use when: Deep dive into object structure and dependencies

---

Space Discovery (3)

get_space_details

Purpose: Get space configuration and settings Returns: Storage allocation, members, permissions, priority Use when: Understanding space setup, auditing configuration

list_space_objects

Purpose: List all objects in a space Returns: Tables, views, models, flows by type Use when: Discovering space content, cataloging objects

get_space_permissions

Purpose: Review space access controls Returns: User permissions, role assignments Use when: Auditing security, understanding access levels

---

Database User Management (5)

create_database_user

Purpose: Create new database user Parameters: Username, password, permissions Use when: Setting up technical users, ETL accounts

update_database_user

Purpose: Modify database user properties Parameters: Username, new permissions Use when: Changing user permissions, updating configuration

delete_database_user

Purpose: Remove database user Parameters: Username Use when: Deprovisioning users, cleanup

reset_database_user_password

Purpose: Reset database user password Parameters: Username Returns: New password (encrypted) Use when: Password recovery, security incidents

list_database_users

Purpose: List all database users Returns: Usernames, permissions, status Use when: Auditing users, permission review

---

Metadata Tools (4)

get_table_metadata

Purpose: Get table structure and properties Returns: Columns, data types, keys, partitions Use when: Designing integrations, understanding table structure

get_view_metadata

Purpose: Get view definition and dependencies Returns: SQL definition, source tables, columns Use when: Understanding view logic, tracing lineage

get_model_metadata

Purpose: Get analytic model structure Returns: Dimensions, measures, associations, variables Use when: Building SAC stories, understanding model capabilities

describe_entity

Purpose: Get comprehensive entity information Returns: Full metadata, relationships, lineage Use when: Detailed analysis, documentation generation

---

Analytical Consumption (4)

query_analytical_data

Purpose: Query analytic models for consumption Parameters: Model name, filters, dimensions, measures Returns: Aggregated data Use when: Testing analytic models, validating calculations

get_analytic_model_structure

Purpose: Get analytic model definition Returns: Model structure, available dimensions/measures Use when: Building queries, understanding model capabilities

execute_smart_query

Purpose: Run SQL queries with aggregation support Parameters: SQL statement Returns: Query results with automatic aggregation Use when: Ad-hoc analysis, testing transformations

preview_data

Purpose: Sample data from tables/views Parameters: Object name, row limit Returns: Sample rows Use when: Quick data inspection, validation

---

ETL-Optimized Relational (4)

query_relational_entity

Purpose: Query relational data efficiently Parameters: Entity name, filters Returns: Relational data without aggregation Use when: ETL operations, data extraction

execute_query

Purpose: Run SQL queries via OData Parameters: SQL statement Returns: Query results Use when: Standard SQL operations, data retrieval

bulk_extract

Purpose: Extract large datasets Parameters: Source, filters, batch size Returns: Data in batches Use when: Migrating data, large-scale exports

stream_data

Purpose: Stream data for real-time processing Parameters: Source, streaming configuration Returns: Data stream Use when: Real-time pipelines, continuous processing

---

Search Tools (2)

semantic_search

Purpose: AI-powered semantic search across catalog Parameters: Natural language query Returns: Relevant objects ranked by relevance Use when: Finding objects by description, fuzzy search

quick_find

Purpose: Fast object lookup by name Parameters: Object name (exact or partial) Returns: Matching objects Use when: Quick navigation, autocomplete scenarios

---

Setup Requirements

Configure environment variables (create .env file):

DATASPHERE_BASE_URL=https://your-tenant.eu10.hcs.cloud.sap
DATASPHERE_CLIENT_ID=your-oauth-client-id
DATASPHERE_CLIENT_SECRET=your-oauth-client-secret
DATASPHERE_TOKEN_URL=https://your-tenant.authentication.eu10.hana.ondemand.com/oauth/token

Getting OAuth Credentials

1. Go to SAP BTP Cockpit → Subscriptions → SAP Datasphere
2. Navigate to Security → OAuth Clients
3. Click Create OAuth Client
4. Configure:
   • Grant Type: Client Credentials
   • Scopes: READ, WRITE (based on needs)
5. Save Client ID and Client Secret
6. Use token URL from subscription details

Authentication

Type: OAuth 2.0 Client Credentials Token Refresh: Automatic (60 seconds before expiration) Storage: Encrypted with Fernet encryption Permissions: Based on OAuth scope (READ, WRITE, ADMIN, SENSITIVE)

Usage Examples

Test Connection

Use MCP tool test_connection to verify Datasphere tenant connectivity

Result: Connection OK, latency 45ms, tenant version 2025.24

Query Data

Use MCP tool execute_smart_query with SQL: SELECT REGION, SUM(REVENUE) FROM fact_sales GROUP BY REGION

Result: Aggregated sales by region

Create Database User

Use MCP tool create_database_user to create ETL_READER with READ permissions

Result: User created, credentials returned

Find Column

Use MCP tool find_assets_by_column to find all objects containing CUSTOMER_ID

Result: List of tables/views with CUSTOMER_ID column

Preview Data

Use MCP tool preview_data from table dim_product with limit 10

Result: First 10 rows of product dimension

Performance

Operation Type	Response Time	Notes
Metadata queries	< 100ms	Cached
Catalog queries	100-500ms	Indexed search
OData queries	500-2,000ms	Depends on data volume
Batch operations	Varies	Up to 50,000 records

Security Features

• ✅ Encrypted Token Storage - Fernet encryption
• ✅ Permission Levels - READ, WRITE, ADMIN, SENSITIVE
• ✅ User Consent - Prompts for high-risk operations
• ✅ Input Validation - SQL injection prevention
• ✅ PII Redaction - Automatic credential/PII masking

Error Handling

Common errors and solutions:

Authentication Failed:

• Verify OAuth credentials in .env
• Check token URL matches tenant region
• Ensure scopes are correct

Connection Timeout:

• Check DATASPHERE_BASE_URL is correct
• Verify network connectivity
• Check firewall/proxy settings

Permission Denied:

• Verify OAuth scope includes required permissions
• Check user has access to requested space
• Review object-level permissions

Integration with Agents

datasphere-modeler:

• Uses: execute_smart_query, query_analytical_data, preview_data, get_*_metadata
• For: Validating designs, testing queries, discovering assets

datasphere-integration-advisor:

• Uses: test_connection, list_spaces, get_tenant_info, search_catalog
• For: Verifying connectivity, discovering spaces, finding objects

datasphere-admin-helper:

• Uses: create/update/delete_database_user, get_current_user, list_spaces
• For: Managing users, auditing permissions, space management

Additional Resources

• GitHub: https://github.com/MarioDeFelipe/sap-datasphere-mcp
• NPM Package: @mariodefe/sap-datasphere-mcp
• Blog Post: https://pub.towardsai.net/sap-datasphere-mcp-server-release-blog-0e82ea33db16
• SAP Datasphere Help: https://help.sap.com/docs/SAP_DATASPHERE

Troubleshooting

Run diagnostics:

npx @mariodefe/sap-datasphere-mcp --test

Enable debug mode (add to .env):

LOG_LEVEL=debug

Check MCP server status:

# From Claude Code, use:
/mcp

See Also

Real-World Use Cases: For comprehensive examples of how these MCP tools are used in practice, see:

• mcp-use-cases.md - 8 detailed use cases demonstrating:
  1. The Monday Morning Health Check - Operations monitoring (45 min/day saved)
  2. Data Lineage Inspection - Impact analysis (3 hours saved)
  3. Pre-Analytics Data Quality Audit - Data validation (2 hours saved)
  4. The Onboarding Speedrun - New hire onboarding (2 days saved)
  5. The Marketplace Shopping Spree - Content discovery (1 hour saved)
  6. The Security Audit - Compliance review (4 hours saved)
  7. The Performance Troubleshooter - Performance debugging (1.5 hours saved)
  8. Cross-Functional Collaboration - Team collaboration (3 hours saved)

Total ROI: $159,100+/year for mid-sized teams

Each use case includes:

• Persona and role
• Quantified time savings
• Complete conversational workflow
• MCP tools used
• Expected results
• ROI analysis