Name: migrate
Author: asabocinski

Prompt file imported from asabocinski/bravo-edd (.codex/prompts/migrate.md). Copyright stays with the author.

Migrate Command

Migrate legacy ETL (stored procedures, SSIS, Informatica) to modern stack

Usage

/migrate <description-or-file>

Examples

/migrate "Convert stored procedures to dbt models"
/migrate legacy/etl_orders_proc.sql
/migrate "Migrate Informatica workflows to Airflow + dbt"
/migrate "Move SSIS packages to Spark + Iceberg"

What This Command Does

Analyzes legacy ETL code or description
Invokes the appropriate agent based on target stack
Loads KB patterns from dbt, spark, and airflow domains
Generates:
- Equivalent modern code (dbt models, PySpark jobs, Airflow DAGs)
- Migration mapping (source → target)
- Data validation queries to compare old vs new output
- Rollback strategy documentation

Agent Delegation

Agent	Role
`dbt-specialist`	Primary — stored proc → dbt model conversion
`spark-engineer`	Primary — heavy ETL → PySpark conversion
`pipeline-architect`	Escalation — orchestration migration
`sql-optimizer`	Escalation — query optimization during migration

KB Domains Used

dbt — model patterns, incremental strategies, macros
spark — PySpark patterns, DataFrame transformations
airflow — DAG patterns, operator selection
sql-patterns — cross-dialect SQL translation

Output

The agent generates modern code equivalents, a migration checklist, and validation queries to ensure data parity.

Install migrate and it becomes one more candidate the router can pick - when it fits.

A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.

{ "prompt": "# Migrate Command\n\n> Migrate legacy ETL (stored procedures, SSIS, Informatica) to modern stack\n\n## Usage\n\n```bash\n/migrate <description-or-file>\n```\n\n## Examples\n\n```bash\n/migrate \"Convert stored procedures to dbt models\"\n/migrate legacy/etl_orders_proc.sql\n/migrate \"Migrate Informatica workflows to Airflow + dbt\"\n/migrate \"Move SSIS packages to Spark + Iceberg\"\n```\n\n---\n\n## What This Command Does\n\n1. Analyzes legacy ETL code or description\n2. Invokes the appropriate agent based on target stack\n3. Loads KB patterns from `dbt`, `spark`, and `airflow` domains\n4. Generates:\n - Equivalent modern code (dbt models, PySpark jobs, Airflow DAGs)\n - Migration mapping (source → target)\n - Data validation queries to compare old vs new output\n - Rollback strategy documentation\n\n## Agent Delegation\n\n| Agent | Role |\n|-------|------|\n| `dbt-specialist` | Primary — stored proc → dbt model conversion |\n| `spark-engineer` | Primary — heavy ETL → PySpark conversion |\n| `pipeline-architect` | Escalation — orchestration migration |\n| `sql-optimizer` | Escalation — query optimization during migration |\n\n## KB Domains Used\n\n- `dbt` — model patterns, incremental strategies, macros\n- `spark` — PySpark patterns, DataFrame transformations\n- `airflow` — DAG patterns, operator selection\n- `sql-patterns` — cross-dialect SQL translation\n\n## Output\n\nThe agent generates modern code equivalents, a migration checklist, and validation queries to ensure data parity.", "variables": [], "notes": "Legacy ETL migration — delegates to dbt-specialist and spark-engineer agents", "metadata": { "source": { "provider": "github-codex-prompts", "repository": "https://github.com/asabocinski/bravo-edd", "path": ".codex/prompts/migrate.md", "ref": "ae4c539580470e782ac67d2dc4a4eabe9fc68336", "url": "https://github.com/asabocinski/bravo-edd/blob/ae4c539580470e782ac67d2dc4a4eabe9fc68336/.codex/prompts/migrate.md", "key": "asabocinski/bravo-edd/.codex/prompts/migrate.md" } } }

migrate

About

Migrate Command

Usage

Examples

What This Command Does

Agent Delegation

KB Domains Used

Output

Install migrate and it becomes one more candidate the router can pick - when it fits.

Use it

Manifest

Reviews