We should use repository rulesets to prevent the deletion or modification of released tags and critical release branches

We can setup rules for released tags and also other branches. See our rules:

https://github.com/apache/cloudstack/rules/

Not 100% sure but seems like anyone with write access can delete releases and checkout the pictures attached

See Apache Flume they have tag protection setup in their `.asf.yaml`:

https://github.com/apache/logging-flume/blob/7b41e071caf1566bf73add2f40b530875d61da10/.asf.yaml#L94

---

Yes, an Apache Software Foundation (ASF) project should use repository rulesets to prevent the deletion or modification of released tags and critical release branches.

### Benefits of Using Rulesets for Releases
* **Supply Chain Security:** Restricting deletions and updates on tags prevents malicious or accidental tampering with published software artifacts.
* **Flexibility and Targeting:** Unlike legacy branch protection, [GitHub Rulesets](https://github.com) can target tags using naming patterns (e.g., `v*` or specific release tags) alongside branches.
* **Audit Transparency:** Anyone with read access can view active rulesets, helping project auditors verify compliance and governance without requiring admin privileges.
* **Preventing Force Pushes:** Rulesets allow projects to block force-pushes and restrict deletions to designated release managers or PMC members.

### Recommended Practices
* Set rulesets to **Active** enforcement for any matching patterns of released tags or stable maintenance branches.
* Restrict bypass permissions strictly to trusted release officers or infrastructure administrators.

Start by reviewing the repository rules linked in the issue and the tag-protection example in .asf.yaml from Apache Flume. Confirm which released tags and critical release branches need protection, then configure active rulesets that block deletion and modification while limiting bypass access. Done means the matching rules and permitted bypass roles are documented and verified.

Written by the indexing model from the issue text.