From a7424c4d80ecfa9197ddd44889bc8ab78a40ea6f Mon Sep 17 00:00:00 2001 From: Alex Nastetsky Date: Tue, 2 Jun 2026 14:59:39 -0400 Subject: [PATCH 1/2] always generate migration report. use subagents with multiple workloads. --- skills/databricks-serverless-migration/SKILL.md | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/skills/databricks-serverless-migration/SKILL.md b/skills/databricks-serverless-migration/SKILL.md index bac1dcd..13b083d 100644 --- a/skills/databricks-serverless-migration/SKILL.md +++ b/skills/databricks-serverless-migration/SKILL.md @@ -623,6 +623,14 @@ Submission is optional and opt-in. We use reports to prioritize new patterns and **Never transmit the report automatically.** The user owns their data and must review before sharing. +## Success report + +If the migration was successful, generate a report as well, using similar methods. For each workload migrated, show what was done. + +## Multiple workload migration + +When migrating multiple workloads in one session, parallelize this work as much as possible to migrate multiple workloads in parallel, including using subagents. + ## Reference Guides For detailed workarounds and code examples beyond the quick fixes above: From 41c551c755019d33b251c64ea4d23f542368060a Mon Sep 17 00:00:00 2001 From: Alex Nastetsky Date: Thu, 4 Jun 2026 10:17:42 -0400 Subject: [PATCH 2/2] adding list of questions to the user --- skills/databricks-serverless-migration/SKILL.md | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/skills/databricks-serverless-migration/SKILL.md b/skills/databricks-serverless-migration/SKILL.md index 13b083d..4e9c998 100644 --- a/skills/databricks-serverless-migration/SKILL.md +++ b/skills/databricks-serverless-migration/SKILL.md @@ -21,6 +21,17 @@ Analyze existing Databricks code for serverless compute compatibility and guide - Troubleshooting serverless-specific errors after migration - Choosing between Performance-Optimized and Standard mode +## Input to get from the user +Ask the user for explicit input parameters for the following items, providing reasonable defaults: +- Which workspace contains the source workloads? +- Which workspace will contain the migrated workloads? +- What is the catalog and schema where output from the test migrated jobs should go? +- How should the test jobs be prefixed or tagged? +- Where in the workspace should the converted code go? +- What compute should be used for reading output produced by classic jobs for comparison (if serverless is not available)? + +Prior to presenting these questions, make sure there are 6 questions (so you don't miss any of the above). + ## Understanding Migration Blockers Migration blockers fall into three categories. Focus your effort on category 2 — that's where this skill helps most.