Treasure Data skills for Claude Code to enhance productivity with TD-specific tools and workflows.
Skills are folders of instructions and resources that Claude loads dynamically to improve performance on specialized tasks. These TD skills teach Claude how to use our tools and follow our best practices.
- sql-skills/trino - Write and optimize SQL queries for Trino with TD best practices
- sql-skills/hive - Create efficient Hive queries following TD conventions
- sql-skills/trino-optimizer - Optimize slow Trino queries, fix timeouts and memory errors, reduce costs
- sql-skills/trino-to-hive-migration - Convert Trino queries to Hive to resolve memory errors and handle large datasets
- sql-skills/trino-cli - Use Trino CLI for interactive queries, data exploration, and terminal-based workflows with TD
- sql-skills/td-mcp - Connect Claude Code to TD via MCP server for natural language data exploration and queries
- realtime-skills/activations - Query activation logs to check for errors and view volume for digital marketing activations
- realtime-skills/identity - Query id change logs to get information about new, updated and merged realtime profiles
- workflow-skills/digdag - Design and implement Treasure Workflow with proper error handling
- workflow-skills/workflow-management - Manage, debug, and optimize existing Treasure Workflows
- workflow-skills/dbt - Use dbt (data build tool) with TD Trino, includes setup, TD-specific macros, and incremental models
- sdk-skills/javascript - Import data to TD using the JavaScript SDK for browser-based event tracking and page analytics
- sdk-skills/python - Query and import data using pytd (Python SDK) for analytical workflows, pandas integration, and ETL pipelines
- tdx-skills/tdx-basic - Core tdx CLI operations for managing TD from command line: databases, tables, queries, and context management
- tdx-skills/parent-segment - Manage CDP parent segments with YAML-based configuration for master tables, attributes, and behaviors
- tdx-skills/segment - Manage CDP child segments with rules, activations, and folder organization
- field-agent-skills/deployment - Best practices for developing, testing, and deploying production-ready Field Agents including R&D workflows and release management
- field-agent-skills/documentation - Comprehensive templates and guidelines for documenting Field Agents with standardized structure, system prompts, and tool specifications
- field-agent-skills/visualization - Professional Plotly visualization best practices with TD color palette, chart specifications, and formatting standards for executive-ready visualizations
- template-skill - Template for creating new TD-specific skills
-
Register the TD skills marketplace:
/plugin marketplace add https://github.com/treasure-data/td-skills -
Browse and install plugins:
Select "Browse and install plugins" from the menu, then choose from:
sql-skills- Trino and Hive query assistance, Trino CLI, and TD MCP serverrealtime-skills- Query activation logs for digital marketing activationsworkflow-skills- Treasure Workflow creation, management, and dbt transformationssdk-skills- TD JavaScript SDK and pytd Python SDKtdx-skills- tdx CLI for managing TD from command linefield-agent-skills- Field Agent deployment, documentation, and visualization best practicestemplate-skill- Template for creating new skills
-
Or install directly:
/plugin install sql-skills@td-skills /plugin install realtime-skills@td-skills /plugin install workflow-skills@td-skills /plugin install sdk-skills@td-skills /plugin install tdx-skills@td-skills /plugin install field-agent-skills@td-skills
Once installed, explicitly reference skills using the skill keyword to trigger them:
"Use the Trino skill to extract data from sample_datasets.nasdaq table"
"Use the Hive skill to write a query for daily user aggregation"
"Use the Trino CLI skill to help me connect to TD from the terminal"
"Use the TD MCP skill to set up Claude Code to access my TD databases"
"Use the activations skill to query activation logs for parent segment 394649"
"Use the digdag skill to create a workflow that runs every morning"
"Use the workflow-management skill to debug this failing workflow"
"Use the dbt skill to create an incremental model for user events"
"Use the JavaScript SDK skill to implement event tracking on my website"
"Use the pytd skill to query TD from Python and load results into pandas"
"Use the tdx-basic skill to list all databases in the Tokyo region"
"Use the parent-segment skill to configure a CDP parent segment"
"Use the segment skill to create child segments with activation rules"
"Use the deployment skill to set up a production publishing workflow"
"Use the documentation skill to create comprehensive Field Agent documentation"
"Use the visualization skill to create a Plotly chart with TD colors"
Tips for triggering skills:
- Include the skill name (Trino, Hive, Trino CLI, TD MCP, activations, digdag, dbt, JavaScript SDK, pytd, tdx, tdx-basic, deployment, documentation, visualization)
- Use the word "skill" in your request
- Be specific about what you want to accomplish
To add a new TD-specific skill:
-
Create a new directory under the appropriate category:
mkdir sql-skills/your-skill-name # or mkdir workflow-skills/your-skill-name # or mkdir field-agent-skills/your-skill-name
-
Create a
SKILL.mdfile with YAML frontmatter (see template-skill) -
Update
.claude-plugin/marketplace.jsonto register the new skill:{ "name": "sql-skills", "skills": [ "./sql-skills/trino", "./sql-skills/hive", "./sql-skills/your-skill-name" // Add your skill path here (must start with ./) ] }
To contribute a new skill or improve an existing one:
- Create or update the skill in a feature branch
- Test the skill with Claude Code
- Submit a pull request with clear documentation
For questions or issues:
- Open an issue in this repository
- Contact the Data Engineering team
Note: These skills are for internal TD use only.