Integrates with SAP Analytics Cloud REST API, providing tools for content management, data export/import, multi actions, calendar, content transport, user management, monitoring, schedule & publication, translation, and smart query.
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_analytics_cloud_mcp list all stories in my SAC tenant"
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.
An MCP (Model Context Protocol) server that exposes the SAP Analytics Cloud REST API as 90 tools consumable by any MCP-compatible client (Custom Agent, Claude Desktop, Claude Code, Cursor, etc.).
The server authenticates against an SAC tenant using OAuth 2.0 Client Credentials and manages CSRF tokens and session cookies automatically. It then exposes the full SAC REST API surface as discrete MCP tools grouped into 11 service areas:
Module | Tools | Covers |
Content | 10 | Stories, Resources, File Repository, Repositories, Widget Query |
Data Export | 22 | Namespaces, Providers, Subscriptions, FactData, MasterData, AuditData, Aggregation, Difference, Currency/Unit tables, Public Dimensions |
Data Import | 24 | Models, Import Jobs lifecycle (create/upload/validate/run/status), One-click Import, Public Dimensions, Currency Conversions, Unit Conversions |
Multi Actions | 2 | Execute and poll status |
Calendar | 3 | Get, update, copy calendar events |
Content Transport | 7 | Export/import jobs, packages, permissions |
User Management | 10 | SCIM v2 Users, Teams/Groups, Bulk operations |
Monitoring | 2 | Audit activity export, model monitoring |
Schedule & Publication | 4 | Schedule CRUD |
Translation | 4 | Artifact metadata, XLIFF download (single + bulk) |
Smart Query | 1 | Intelligent routing (Relational vs Analytical) |
Plus a ping connectivity check tool (90 total), and Prompt support for guided workflows.
Related MCP server: SAP OData to MCP Server
Node.js >= 18.0 (required for native fetch and Headers.getSetCookie)
npm
An SAP Analytics Cloud tenant with an OAuth client (see below).
Open your SAC tenant and go to System > Administration > App Integration > Add a New OAuth Client.
Set the purpose to Interactive Usage and API Access.
Note down the following four values:
Base URL -- https://<tenant>.<datacenter>.sapanalytics.cloud
Token URL -- shown on the OAuth client detail page
Client ID
Client Secret
git clone https://github.com/JumenEngels/sap_analytics_cloud_mcp
cd sap_analytics_cloud_mcp
npm install
npm run buildThe easiest way to test the server locally (after doing mcp setup above) is using the included Agent Client. It connects to the server and uses an LLM (SAP AI Core, Anthropic, OpenAI, etc.) to interact with your SAC tenant.
Navigate to the client directory and run the setup script:
cd AgentClient
npm install
npm run setupThis creates mcp_agentclient.json. Open this file and fill in your credentials:
SAC Credentials: Fill in sac section with the values from "SAC Tenant Setup".
LLM Credentials: Fill in aicore, anthropic, or openai sections.
Default Provider: Set "defaultProvider" to your preferred LLM (e.g., "genaicore", "anthropic").
npm run build
npm startThe client will automatically load all necessary configuration (both for the LLM and the SAC Server) from your JSON file.
If you are using Claude Desktop, Cursor, or another MCP client, you need to configure the server environment variables directly.
The server reads four required environment variables at runtime:
Variable | Description | Example |
| Tenant root URL (no trailing slash) |
|
| OAuth token endpoint |
|
| OAuth Client ID |
|
| OAuth Client Secret |
|
To automatically generate a configuration file with the correct absolute path:
npm run setupThis creates an mcp_config.json file in the project root. fill in your SAC credentials there and copy the content to your Claude Desktop config file.
Add the server with the command node build/index.js and set the 4 environment variables in the client's configuration UI.
The smart_query tool simplifies data access by intelligently routing your request:
Relational: Simple SELECT * queries are routed to the OData API for raw entity data.
Analytical: Queries with aggregations (SUM, COUNT, GROUP BY) are analyzed and routed to the Widget Query API.
SELECT SUM(Amount) FROM Model123 GROUP BY RegionThe server provides pre-defined prompts to guide users through common workflows:
explore_content: Discover stories and files in your tenant.
analyze_story: Deep dive into a specific story's metadata and dependencies.
system_health_check: Quick status check of connectivity and recent failures.
docker build -t sac-mcp-server .
docker run -e SAC_BASE_URL="..." ... sac-mcp-serverWatch mode recompiles on every file change:
npm run devWrite Operations: All tools that modify data require allowalteration=true argument.
Returns pong — useful for verifying connectivity
Inputs: no named parameters.
Export audit activity logs with OData query params.
Inputs: $top, $skip, $filter, $select, +1 more.
Copy/create a calendar event.
Inputs: body (required), allowalteration.
Get a calendar event by ID.
Inputs: eventId (required).
Update a calendar event (partial).
Inputs: body (required), eventId (required), allowalteration.
Get OData metadata for admin service.
Inputs: $format.
Create data export subscription for delta tracking.
Inputs: ProviderID (required), NamespaceID (required), EntitySetName (required), Filter, +4 more.
Delete data export subscription(s).
Inputs: providerId (required), namespaceId (required), deleteChain, subscriptionId, +2 more.
Get audit data (change history). Requires Data Audit enabled on model.
Inputs: provider (required), namespace (required), $top, $skip, +5 more.
Get unit conversion rates. Namespace: sac_unit_conversion.
Inputs: provider (required), $top, $skip, $filter, +3 more.
Get currency conversion rates. Namespace: sac_currency_tables.
Inputs: provider (required), $top, $skip, $filter, +3 more.
Generic: get data from any provider entity set. For Distinct queries use '{col}___Distinct'.
Inputs: provider (required), entitySet (required), namespace (required), $top, +11 more.
Get master data for a single dimension ({DimensionName}Master entity set).
Inputs: provider (required), dimension (required), namespace (required), $top, +5 more.
Get master data with parent-child hierarchy for a dimension.
Inputs: provider (required), dimension (required), namespace (required), $top, +5 more.
Get fact data. Supports OData params, delta tracking (trackChanges/deltaid/externalid), CSV export.
Inputs: provider (required), namespace (required), $top, $skip, +10 more.
Get aggregated fact data grouped by dimensions and measures. PREREQUISITE: call sac_import_get_model_metadata first to discover valid dimension and measure names for $select. Use $select for dims/measures. 'provider' is the model ID — resolve name→ID via sac_filerepository_list ($filter=name eq 'MODEL_NAME' and resourceType eq 'ANALYTIC_MODEL'). Namespace for analytic models is 'sac'.
Inputs: $select (required), provider (required), namespace (required), $top, +7 more.
Get measure value differences over subscription period. Requires trackChanges or deltaid.
Inputs: provider (required), namespace (required), $top, $skip, +5 more.
Get master data and level-based hierarchies for all dimensions.
Inputs: provider (required), namespace (required), $top, $skip, +5 more.
Get namespace details.
Inputs: namespaceId (required).
List available entity sets for a provider.
Inputs: provider (required), namespace (required).
Get OData metadata (EDMX) for a provider.
Inputs: provider (required), namespace (required), $format.
Get public dimension master data. Namespace: sac_public_dimensions.
Inputs: provider (required), $top, $skip, $filter, +3 more.
Get subscription by composite key.
Inputs: providerId (required), namespaceId (required), subscriptionId (required).
List data export namespaces (sac, sac_currency_tables, sac_public_dimensions, sac_unit_conversion).
Inputs: no named parameters.
List data providers (models) in a namespace.
Inputs: namespaceId (required).
List subscriptions for a provider.
Inputs: providerId (required), namespaceId (required).
List all data export subscriptions.
Inputs: no named parameters.
List story types, enumerate content types, and filter file repository resources via GET /api/v1/filerepository/Resources. Use to list all story types (standard story, story template, analytic application), enumerate design experiences (Classic vs Optimized), or filter stories/apps by user, date, or type. OData $filter, $top, $orderby, $select, $count all work. IMPORTANT — visibility scope: By default (applyManagePrivilege=false), only content the service account has Read access to is returned (typically a small subset of tenant content). Set applyManagePrivilege=true to get ALL tenant content (all users' private folders, public folders, workspaces) — requires the service account to have the "Manage" permission for Private Files and Public Files. Always use applyManagePrivilege=true when the user wants to search/list across the whole tenant. Response fields: resourceId, objectId, name, description, resourceType, resourceSubtype, createdTime, createdBy, modifiedTime, modifiedBy, folderType, workspaceId, workspaceName, openURL, isMobile, isFeatured. Filterable fields: resourceType (values: STORY, APPLICATION, DATAACTION, PLANNINGSEQUENCE, MULTIACCOUNT, DIMENSION, ANALYTIC_MODEL), createdBy (exact username/ID), modifiedBy, createdTime, modifiedTime, name, folderType (PUBLIC, PRIVATE, SYSTEM, INPUT_SCHEDULE). STORY TYPES — resourceSubtype and objectId encoding: resourceSubtype="" (empty) → standard story (covers both Classic and Optimized design experience) resourceSubtype="TEMPLATE" → story template resourceSubtype="APPLICATION" → Analytic Application (scripted, code-based) Design experience is NOT in resourceSubtype — it is encoded in the objectId prefix: objectId prefix "t.D:" → Classic design experience (legacy) objectId prefix "t.B:" → Optimized design experience (recommended, newer) To list story types: $select=name,resourceSubtype,objectId and $orderby=resourceSubtype asc — groups stories by subtype and lets you read the objectId prefix for design experience. Filter examples: $filter=resourceType eq 'STORY' — all stories $filter=resourceType eq 'STORY' and createdBy eq 'USERNAME' — stories by a specific user $filter=resourceType eq 'STORY' and modifiedBy eq 'USERNAME' — stories last modified by user $filter=resourceType eq 'STORY' and createdTime gt 2024-01-01T00:00:00Z — recently created WORKSPACE / TEAM FILES: To list all workspaces (team files spaces), filter by folderType eq 'INPUT_SCHEDULE' and workspaceId ne null. This returns exactly one entry per workspace with its workspaceId and workspaceName. Example: $filter=folderType eq 'INPUT_SCHEDULE' and workspaceId ne null with $select=workspaceId,workspaceName and $orderby=workspaceName asc.
Inputs: $top, $skip, $count, $filter, +3 more.
Create import job for a currency conversion table.
Inputs: currencyConversionId (required), Mapping, importType, JobSettings, +1 more.
Create import job for a model. Returns JobID/JobURL.
Inputs: modelId (required), Mapping, importType, JobSettings, +1 more.
Create import job for a public dimension.
Inputs: publicDimensionId (required), Mapping, allowalteration.
Create import job for a unit conversion table.
Inputs: unitConversionId (required), Mapping, importType, JobSettings, +2 more.
Delete an import job.
Inputs: jobId (required), allowalteration.
Get currency conversion import types and settings.
Inputs: currencyConversionId (required).
Get currency conversion metadata: columns, types, keys.
Inputs: currencyConversionId (required).
Get rejected rows with rejection reasons.
Inputs: jobId (required).
Get job status: READY_FOR_DATA, READY_FOR_WRITE, PROCESSING, COMPLETED, FAILED.
Inputs: jobId (required).
Get model import types, versions, settings, and metadata URL.
Inputs: modelId (required).
Get model dimension names, measure names and field types. Call this BEFORE sac_export_get_fact_data_aggregation to discover valid $select fields (e.g. PD_O_BusinessUnit, SignedData). Required whenever field names are unknown. Use when: aggregating fact data, exporting by dimension, querying model structure.
Inputs: modelId (required).
Get public dimension import types and metadata URL.
Inputs: publicDimensionId (required).
Get public dimension metadata: columns, types, keys.
Inputs: publicDimensionId (required).
Get unit conversion import types and settings.
Inputs: unitConversionId (required).
Get unit conversion metadata: columns, types, keys.
Inputs: unitConversionId (required).
List currency conversion tables for import.
Inputs: top.
List all import jobs.
Inputs: top.
List models available for data import. Does NOT support name filtering — returns all models. To find a model ID by name, use sac_filerepository_list with $filter=name eq 'MODEL_NAME' and resourceType eq 'ANALYTIC_MODEL' — the returned resourceId is the model ID.
Inputs: top.
List public dimensions available for import.
Inputs: top.
List unit conversion tables for import.
Inputs: top.
One-click: create job, upload, validate, run in one request.
Inputs: Data (required), modelId (required), Mapping, importType, +3 more.
Execute validated import job.
Inputs: jobId (required), allowalteration, overrideExecuteWithFailedRows.
Upload data to an import job. Returns upserted/failed row counts.
Inputs: Data (required), jobId (required), Mapping, DeletedData, +2 more.
Validate import job before running.
Inputs: jobId (required).
Get OData metadata document for the SAC REST API.
Inputs: no named parameters.
Get monitoring data for a model.
Inputs: modelId (required).
Execute a multi action. ID format: :. Body requires parameterValues array (empty if none).
Inputs: multiActionId (required), allowalteration, parameterValues.
Get execution status of a multi action.
Inputs: executionId (required), multiActionId (required).
List repositories. Supports OData query params.
Inputs: $top, $skip, $expand, $filter, +4 more.
Get a single resource by ID.
Inputs: resourceId (required).
List resources via GET /api/v1/Resources (OData v2, returns XML/Atom format — not recommended for filtering). Use sac_filerepository_list for JSON responses with working OData $filter support.
Inputs: $top, $skip, $expand, $filter, +4 more.
Create a publication schedule.
Inputs: body (required), allowalteration.
Delete a publication schedule.
Inputs: scheduleId (required), body, allowalteration.
Get schedule creation template with defaults.
Inputs: no named parameters.
Update a publication schedule (partial).
Inputs: body (required), scheduleId (required), allowalteration.
Copy a story.
Inputs: sourceStoryId (required), newName, copyToFolder, allowalteration.
Delete a story.
Inputs: storyId (required), allowalteration.
List stories via GET /api/v1/stories. WARNING: This endpoint does NOT support OData filtering. All query parameters ($filter, $top, $orderby, etc.) are silently ignored by the SAC API — it always returns every story on the tenant (3000+). The response does NOT include an 'owner' field. Response fields per story: storyId, name, createdBy, changedBy (and optionally model metadata). TO FILTER STORIES BY USER OR TYPE use sac_filerepository_list instead, which supports real OData $filter on createdBy, modifiedBy, resourceType, createdTime, etc.
Inputs: includeModels.
Rename a story.
Inputs: newName (required), storyId (required), allowalteration.
Create a team/group (SCIM v2).
Inputs: body (required), allowalteration, createTeamFolder.
List user security groups via SCIM v2 (/api/v1/scim2/Groups). Returns groups used for role assignments and access control (e.g. T_SALES, T_HR). These are identity/authorization groups, not file storage folders.
Inputs: count, filter, startIndex.
Partial update of a team/group (SCIM v2 PATCH).
Inputs: body (required), groupId (required), allowalteration.
Download XLIFF translation file for an artifact.
Inputs: artifactId (required).
Download XLIFF files for multiple artifacts.
Inputs: body (required).
Get metadata for a translatable artifact.
Inputs: artifactId (required).
List translatable artifacts.
Inputs: $top, $skip, $filter.
Create a content transport export job.
Inputs: body (required), allowalteration.
Create a content transport import job.
Inputs: body (required), allowalteration.
Delete a content item.
Inputs: itemId (required), allowalteration.
Get content item details within a wave.
Inputs: itemId (required), waveItemId (required).
Get content transport job status.
Inputs: jobId (required).
Get permissions for a content item.
Inputs: itemId (required).
Move a content item to a target location.
Inputs: itemId (required), targetId (required), body, allowalteration.
Bulk SCIM operations (create/update/delete multiple users/groups).
Inputs: body (required), allowalteration.
Create a user (SCIM v2).
Inputs: body (required), allowalteration.
Delete a user.
Inputs: userId (required), allowalteration.
Get a user by UUID.
Inputs: userId (required).
List users (SCIM v2).
Inputs: count, filter, startIndex.
Partial update of a user (SCIM v2 PATCH).
Inputs: body (required), userId (required), allowalteration.
Full replace of a user (SCIM v2 PUT).
Inputs: body (required), userId (required), allowalteration.
Get widget data from a story.
Inputs: storyId (required), widgetId (required), type.
Execute a SQL-like query against SAC. Auto-selects OData vs Widget Query based on query type.
Inputs: query (required), model_id, space_id.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 90 tools
Tools are well-organized by domain with clear prefixes (sac_export_, sac_import_, sac_users_, etc.), and descriptions are detailed. However, the sheer number (90 tools) can still cause confusion, especially with multiple tools for similar purposes (e.g., sac_filerepository_list vs sac_stories_list).
All tools follow a consistent pattern: sac_<domain>_<action> with snake_case (e.g., sac_export_get_fact_data, sac_import_create_job). No mixing of conventions.
With 90 tools, the server is extremely large for an MCP server. While it covers many SAP Analytics Cloud features, the number is excessive and overwhelms the typical scope (3-15 tools). Many tools could be consolidated or omitted.
The tool set covers a wide range of SAC operations: data export/import, user and team management, content management, transport, translation, scheduling, monitoring, and more. Minor gaps exist (e.g., no direct story update tool), but overall it is comprehensive.
Query your warehouse or a CSV with Claude/ChatGPT over MCP, governed by table-level ACL + audit.
Official Microsoft MCP Server to query Microsoft Entra data using natural language
Query your org's data in natural language — read-only MCP access to SQL, NoSQL, files & warehouses.
The Ramp MCP server enables users to securely connect Ramp with AI assistants like ChatGPT and Claude to query financial data and take actions using natural language. It transforms Ramp's developer API into a SQL interface that LLMs can query, allowing admins to analyze spend trends, identify cost savings, and run complex SQL analyses on comprehensive datasets (transactions, purchase orders, vendors, users), while all users can manage cards, view transactions, request reimbursements, and get expense policy answers.
