Treasure Data MCP Server
OfficialServer Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| TD_SITE | No | Region endpoint - us01 (default), jp01, eu01, ap02, ap03, dev | us01 |
| TD_API_KEY | Yes | Your Treasure Data API key (required) | |
| TD_DATABASE | No | Default database for queries (e.g., sample_datasets) | |
| TD_ENABLE_UPDATES | No | Enable write operations (execute tool) - false (default), true | 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 | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_databasesA | List all available databases in Treasure Data |
| list_tablesB | List all tables in a specific database |
| describe_tableC | Get column information for a specific table |
| queryA | Execute a read-only SQL query (SELECT, SHOW, DESCRIBE) using Trino SQL dialect with Treasure Data UDFs. For better performance on large tables, use td_interval(time, '-30d/now') to limit the time range. |
| executeB | Execute write operations (UPDATE, DELETE, INSERT, etc.) when enable_updates=true |
| use_databaseA | Switch the current database context for subsequent queries |
| current_databaseA | Get the current database context being used for queries |
| list_parent_segmentsB | [EXPERIMENTAL] List all parent segments (audiences) in Customer Data Platform. No parameters required. |
| get_parent_segmentA | [EXPERIMENTAL] Get details of a specific parent segment by its ID. Requires parent_segment_id parameter. Use list_parent_segments first to find available parent segment IDs. |
| list_segmentsA | [EXPERIMENTAL] List all segments under a specific parent segment. Requires parent_segment_id parameter. Use list_parent_segments first to find available parent segment IDs. |
| list_activationsA | [EXPERIMENTAL] List all activations (syndications) for a specific segment. Requires both parent_segment_id and segment_id parameters. Use list_parent_segments and list_segments first to find available IDs. |
| parent_segment_sqlB | [EXPERIMENTAL] Get the SQL statement for a parent segment (audience). Requires parent_segment_id parameter. Use list_parent_segments first to find available parent segment IDs. |
| segment_sqlA | [EXPERIMENTAL] Get the SQL statement for a segment with filtering conditions applied to the parent segment (audience) SQL. Requires both parent_segment_id and segment_id parameters. Use list_parent_segments and list_segments first to find available IDs. |
| get_segmentA | [EXPERIMENTAL] Get detailed information about a specific segment including its filtering rules. Requires both parent_segment_id and segment_id parameters. Use list_parent_segments and list_segments first to find available IDs. |
| list_projectsC | List all workflow projects |
| list_workflowsC | List workflows. Optionally filter by project name. |
| list_sessionsC | List workflow execution sessions with filtering options |
| get_session_attemptsB | Get detailed information about all attempts for a specific session |
| get_attempt_tasksC | List all tasks within an attempt with their execution status |
| get_task_logsC | Retrieve logs for a specific task |
| kill_attemptB | Request cancellation of a running attempt |
| retry_sessionB | Retry a failed session from the beginning or a specific task |
| retry_attemptC | Retry a specific failed attempt with resume capabilities |
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 23 tools
Most tools target a distinct resource+action pair (list_databases vs list_tables vs describe_table; query vs execute is cleanly split by read-only vs write). The only real confusion candidates are retry_session vs retry_attempt and parent_segment_sql vs segment_sql, but the descriptions explicitly differentiate the scope (session vs attempt, parent vs filtered) so an agent can still choose correctly.
The large majority of tools follow a verb_noun convention (list_*, get_*, retry_*, kill_attempt, describe_table, use_database). A handful deviate: bare verbs 'query' and 'execute', the noun-first getter 'current_database' (which pairs awkwardly with 'use_database'), and noun-first 'parent_segment_sql'/'segment_sql'. Minor inconsistencies, not enough to break predictability.
23 tools is on the heavy side and reflects three loosely bundled products in one server: SQL/database access, experimental CDP segment APIs, and workflow/session/attempt/task inspection. Each tool does earn its place, but the workflow-debugging chain (list_sessions → get_session_attempts → get_attempt_tasks → get_task_logs) plus the experimental tag suggests the surface could be split or trimmed.
Read and write SQL, context switching, and deep workflow introspection are well covered, but there are noticeable gaps: no create/drop/alter table or database operations, no way to fetch a single workflow definition or start/stop a workflow job, and CDP coverage stops at listing/reading (no segment or audience creation). These are recoverable via query/execute for some cases but leave the workflow lifecycle one-sided.