di-agent-query-substrait

Generate Substrait query plans from natural language data requests.

3|Updated May 1, 2026
One-click install
npx skills add https://github.com/IBM/ibm-watsonx-data-integration-skills --skill di-agent-query-substrait
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: di-agent-query-substrait
Source: https://github.com/IBM/ibm-watsonx-data-integration-skills/tree/main/agent/skills/di-agent-query-substrait
Command: npx skills add https://github.com/IBM/ibm-watsonx-data-integration-skills --skill di-agent-query-substrait

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill turns natural language data requests into valid Substrait query plans, reducing the effort needed to author and debug query DSL by hand.

Core Features & Use Cases

  • Generate Substrait DSL for query translation, aggregation, joins, filtering, and plan building.
  • Verify source schemas from project assets before generating a plan so the output matches the actual data.
  • Recover from compile errors by correcting the DSL and, when requested, run the plan or convert it into Elyra pipeline JSON.

Quick Start

Ask the agent to generate a Substrait DSL plan for your data request and include the source table names so it can verify the schema before compiling.

Frequently Asked Questions about di-agent-query-substrait

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate a Substrait query plan from natural language?

To generate a Substrait query plan from natural language, provide your data request and source table names so the agent can verify schemas before compiling the DSL.

Can I convert a Substrait plan into an Elyra pipeline?

Yes, you can convert a Substrait plan into an Elyra pipeline JSON. After generating and compiling the DSL, request Elyra conversion through the available MCP tools.

Does schema validation happen before Substrait DSL generation?

Schema validation occurs before Substrait DSL generation by verifying source schemas from project assets, ensuring the output query plan matches the actual data structure.

How do I fix Substrait DSL compile errors during query translation?

To fix Substrait DSL compile errors, the agent uses a compilation and correction loop that automatically identifies issues and corrects the DSL until it compiles successfully.

What is the best way to author Substrait plans for IBM watsonx.data integration?

The best way to author Substrait plans for IBM watsonx.data integration is using natural language data requests, which the agent translates into valid Substrait DSL for query planning.

Can I execute a Substrait plan after generating it from natural language?

Yes, you can execute a Substrait plan after generation. Once the DSL is compiled and corrected, you can request optional execution of the plan through MCP tools.