Prune — Version & Trash Cleanup for Confluence

Security & data handling

Last updated: 1 November 2026

QuestionAnswer
HostingAtlassian Forge only. Eligible for Runs on Atlassian (confirmed by forge deploy).
External network calls (egress)None. The app manifest declares no external permissions.
Data stored by the appForge storage only: saved rules, trash first-seen dates (content ID + date), job progress, and run summaries for the last 30 runs (counts, content IDs, space IDs). No page content, titles or user names.
Data stored outside AtlassianNone.
Scopesread:space:confluence, read:page:confluence, read:blogpost:confluence (inventory and dry run); delete:content:confluence (delete old versions); delete:page:confluence, delete:blogpost:confluence (purge from trash); read:confluence-user (administrator check); storage:app.
Who can use itConfluence administrators only. The page lives in Confluence administration, and every request is checked on the server: the user's account (from Atlassian's request context) must have the Confluence "administer" permission.
Who the app acts asThe app (Forge app user), so daily rules can run without a browser. Only administrators can start jobs or save rules.
Safety railsDry run by default; typed DELETE confirmation; the current version is never deleted; the version list is re-read before every delete; trash items restored meanwhile are skipped; one job at a time; daily trash rule needs at least 1 day.
Audit trailThe app's history lists the last 30 runs with a CSV report. Confluence's audit log records the deletions as well.
Personal dataNone stored. Version author IDs are read from Confluence only in memory.
Data residencyForge storage follows Atlassian's data residency for Forge apps on your site.
Analytics and trackingNone.
Sub-processorsNone besides Atlassian, which hosts Confluence and Forge.
Vulnerability reportsEmail prune@kenanislamoglu.com with the subject "Security". We acknowledge within 2 business days.