Dataiku MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| DATAIKU_URL | Yes | DSS base URL | |
| DATAIKU_API_KEY | Yes | DSS API key | |
| DATAIKU_PROJECT_KEY | No | Default project key (optional) | |
| DATAIKU_DEBUG_LATENCY | No | Set to '1' or 'true' to include per-tool timing diagnostics | |
| DATAIKU_REQUEST_TIMEOUT_MS | No | Per-attempt request timeout in milliseconds (default: 30000) | |
| DATAIKU_RETRY_MAX_ATTEMPTS | No | Max attempts for retry-enabled requests (GET only, default: 4, cap: 10) |
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
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| projectA | Project ops: list/get/metadata/flow/map. map returns normalized connectivity in structuredContent.map; includeRaw adds original graph payload. |
| datasetB | Dataset ops: list/get/schema/preview/metadata/download/create/update/delete. get is summary-first; set includeDefinition=true to include full JSON definition. |
| recipeC | Recipe ops: list/get/create/update/delete/download. get is summary-first; set includePayload=true to include payload snippets. |
| jobB | Job ops: list/get/log/build/buildAndWait/wait/abort. get is summary-first; set includeDefinition=true to include full JSON definition. |
| scenarioA | Scenario ops: list/run/status/get/create/update/delete. get is summary-first; use includeScript/includeDefinition for details. |
| variableB | Project variable ops: get/set. set merges provided standard/local keys with existing values. |
| managed_folderC | Managed folder ops: list/get/contents/download/upload/delete_file. |
| connectionA | Connection discovery (action: infer). Default mode=fast uses DSS connection names; mode=rich scans project datasets for inferred type/schema/managed details. |
| code_envB | Code env ops: list/get. get returns package summaries; set full=true for full package lists. |
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 9 tools
Each tool maps to a distinct Dataiku resource type (project, dataset, recipe, job, scenario, variable, managed_folder, connection, code_env), with no overlapping responsibilities. The descriptions clearly differentiate the scope of each tool.
All tool names are single lowercase nouns or snake_case compound nouns (e.g., managed_folder, code_env), following a predictable resource-based pattern. No mixing of styles or vague verbs.
With 9 tools, the server is well-scoped for Dataiku's domain. Each entity earns its place and the count is comfortably within the ideal 3-15 range.
The core workflows (list/inspect datasets, recipes, jobs, scenarios; run builds and scenarios) are well covered. Some gaps exist, such as project creation/deletion and code_env management, but these are secondary for common agent tasks.