Run a Glue locally for development and testing before deploying. This is the fastest way to iterate, the dev command sets up your Glue locally and tunnels events to your local machine. To make testing your Glues fast:
console logs will get printed to your terminal
your code is hot reloaded as you make changes, the next event that gets triggered will run your updated code
you can replay past events to your production glue to test your new code against real data using the --replay flag
you can replay the last event you received to test your new code against real data by pressing r in the dev terminal.
you can generate sample events for triggers that support it by pressing s in the dev terminal.
you can attach a debugger to your local Glue process. See debugger options: Debugging
Deploy a Glue so it can run in our hosted environment. Use this when you’re ready for your Glue to run reliably (on schedules, webhooks, or other triggers) in the background.
View execution history and optionally tail live executions. This is your primary tool for debugging in production: confirm that runs are happening, inspect failures, and understand what your Glue did. By default, this will show executions and a few lines of console logs per execution. You can adjust how many executions are shown and how many lines of console logs are shown. You can also filter executions to just the failed ones to aid in debugging.
Inspect any resources by name or ID. This is useful to see the details of a Glue, a Deployment, an Execution, an Account, or a Trigger and usually shows more details that list based commands.
Replay a past execution by ID on your deployed Glue. This is useful when investigating a failure or surprising behavior: you can re-run the exact same execution to reproduce and debug the issue.
Create a new basic Glue that’s ready to run. It can generate an empty scaffold or a fully fleshed out Glue from a natural language description. Use this to get a working starting point quickly instead of bootstrapping a project by hand.
List your deployed Glues. This will show you the name, running status, number of runs, number of errors, last run, last deployed, and tags for each Glue.
Add, remove, or replace tags on one or more Glues. Tags are useful for grouping Glues (by team, customer, purpose, or owner) so you can filter and manage at scale.
Archive one or more Glues. Use this to clean up Glues you don’t want running or showing up in default lists. You can only archive stopped Glues but they can be restored and re-deployed later.
Stop a deployed Glue. Use this as an emergency brake if a Glue is misbehaving, generating noise/cost, or you need to pause it while you investigate. You can only stop running Glues.
Show which user/account the CLI is currently authenticated as. Useful for quickly verifying you’re operating in the right account/environment before deploying or deleting anything.
List all external service accounts connected to Glue. This will show you the service name, the email address or account associated with the service, the type of account (OAuth, API key, etc.), the scopes, and the associated Glues that rely on it.
Delete a connected account by ID. Use this to revoke access, remove stale/incorrect connections, or clean up after testing. You can only delete accounts that are not associated with any running Glues.
Share a Glue file by creating a GitHub secret gist. Use this to send someone a reproducible Glue example (or ask for help) without publishing it publicly.
