Is anyone here actively using labels in dynamic pipelines to generate the entire pipeline? If so, how are you using them? I’m looking for ideas and real-world examples, since there’s very little documentation available both from the community and even from Bitbucket, despite them mentioning internal usage.
For context, we’re trying to move away from shared pipelines and transition to dynamic pipelines that can generate full pipelines for repositories with common steps. It seems like labels might be the right approach to help generate the spec file and pass that information to the Forge app, with the Forge remote app handling the generation of the pipeline based on the label information provided.
The only active post I have found on the subject are the following resources:
- some old thread on the subject
- video on dynamic pipelines
- Atlas Camp 2026
