q re: triggering flow runs: we're mostly re-using

cloud

from

cli.run

to start runs since it handles a lot of the boilerplate of checking status / fetching logs for us. but it doesn't really seem like it was designed for programmatic use: in particular there's no meaningful return value or exit code, so it can't be used to see if a flow run succeeded or failed or anything else about it. would changing that function to return more info (say flow run ID and state) make sense?

just opened a PR with a small change that would resolve our issue: returning the flow run ID from the CLI run methods https://github.com/PrefectHQ/prefect/pull/2242