AWS · Cloud service
AWS Glue with OrchestrAI
Catalog exported 2026-09-02
Run AWS Glue from chat: Data Catalog databases and tables, crawlers, ETL jobs, and job runs.
OrchestrAI exposes 8 Glue operations: 3 are low-risk (read-only or low-impact), and 5 create or modify resources and run only after you confirm the plan. 5 of them also carry a step-level approval gate.
What teams use it for
Data engineering teams use OrchestrAI to create a Data Catalog database, define tables, set up a crawler over an S3 prefix, create an ETL job pointing at a script, and kick off job runs, then browse databases and table schemas. Every create and start operation waits for your confirmation, and catalog reads are low risk. Starting a crawler and checking the status of a job run are not available, so after a job starts you follow progress in the Glue console.
Every Glue operation, with its risk level
| Operation | What it does | Risk | Step-level approval |
|---|---|---|---|
Get Glue Table |
Get Glue table schema | Low risk | No |
List Glue Databases |
List AWS Glue databases | Low risk | No |
List Glue Tables |
List tables in Glue database | Low risk | No |
Create Glue Crawler |
Create AWS Glue crawler for data discovery | Creates resources | Yes |
Create Glue Database |
Create AWS Glue Data Catalog database | Creates resources | Yes |
Create Glue ETL Job |
Create AWS Glue ETL job for data transformation | Creates resources | Yes |
Create Glue Table |
Create Glue table | Creates resources | Yes |
Start Glue Job Run |
Start an AWS Glue ETL job run | Creates resources | Yes |
Risk tiers come from the catalog: low is read-only or low-impact, medium creates resources and is reversible, high modifies existing resources, destructive may lose data. Every plan that creates or changes resources is shown with its cost estimate and waits for your confirmation. Operations marked with a step-level approval pause again on their own step. Destructive operations require a typed risk phrase.
Prompts that work
- Create a Glue database called analytics_raw and a crawler over s3://data-lake/raw/events/
- Show me the schema of the events table in analytics_raw
- Start a run of the nightly-transform Glue job
Before anything runs
Every mutation shows its plan, cost estimate, and blast radius, then waits for your confirmation. Destructive operations require a typed risk phrase. Credentials are minted per run through OIDC federation and discarded afterward; nothing you create here is invisible later, because every resource lands in the desired-state ledger where drift is detected and can be converged. Details on the security page.
Frequently asked questions
- Can OrchestrAI start a Glue crawler?
- No, it can create crawlers, but starting one is not in the operation set, so run it from the console.
- Does OrchestrAI ask before running a Glue ETL job?
- Yes, starting a job run is medium risk with confirmation, so the job name and arguments are shown before it runs.
- Which Glue operations need an extra approval step?
- 5 operations carry a step-level approval gate on top of plan confirmation: Create Glue Crawler, Create Glue Database, Create Glue ETL Job, Create Glue Table, Start Glue Job Run. None of them is classed destructive.
Other AWS services
Related integrations
Try it on your own account
Connect your cloud read-only and see your resources, drift, and costs before anything runs. $5 minimum to start. Unused credits refunded in your first 14 days.
Unused credits refunded in your first 14 days.