Question: GSoC proposal scope for GHCR cleanup workflow

[AnjaliMishra1st]

Hi everyone 👋
I’m currently working on the GHCR cleanup workflow and exploring it further.
I’m planning to write a GSoC proposal based on this work, and I wanted to ask if this would be a suitable topic, or if you’d recommend expanding the scope (for example, broader container image lifecycle management or support for other registries).

I’d really appreciate any guidance on scope and expectations.

Thanks!

[julianocosta89]

Hello @AnjaliMishra1st 👋🏽
What is GSoC?

[AnjaliMishra1st]

Hi @julianocosta89 👋🏽
Apologies for the confusion! By GSoC, I meant Google Summer of Code, which is a program where contributors propose and work on open-source projects with mentorship.
I’m exploring whether extending the GHCR cleanup workflow (or expanding it into a broader container image lifecycle management solution) could be a suitable project idea for a GSoC proposal.

I’d really appreciate your thoughts on whether this scope would be appropriate or if you’d recommend broadening it further. Thanks!

[AnjaliMishra1st]

Hi everyone 👋
I’ve prepared a draft GSoC proposal based on the GHCR cleanup workflow I’ve been working on.
I tried to incorporate earlier feedback, especially around retention strategy, safety, and avoiding aggressive deletion.
I would really appreciate any feedback on:

• Scope of the project
• Cleanup strategy and retention rules
• Any suggestions to improve the proposal
  Proposal link:
  https://docs.google.com/document/d/1xr_jlHvu3HbbZLHeb2C6zyIXsjjEqoeQaaT0wsWRAfU/edit?usp=sharing

Thanks a lot for your time!

[julianocosta89]

@AnjaliMishra1st I'm not aware how GSoC works but I'm happy to help wherever I can.
The doc looks good, I do not understand the timelines though.

Wouldn't that be 1-2 PRs max?

[AnjaliMishra1st]

Hi @julianocosta89 👋🏽

Thanks for the feedback!
You’re right — a basic GHCR cleanup workflow might only result in a small number of PRs. For GSoC, I’m planning to expand this into a broader container image lifecycle management system, including:

• Configurable retention policies
• Safety features (dry-run, logs, restore)
• Multi-registry support
• CI/CD integration and observability
  So the idea is to build a reusable, scalable solution rather than just a single workflow.
  I’m happy to refine the timeline as well. Would love your thoughts on whether multi-registry support or observability would be more valuable to prioritize.
  Thanks again! 🙂