Buying and licensing
Does it slow Jira down? No. The app adds no background indexing. Small triggers run when a work item is created, updated or deleted, and when a comment or a view event fires; each reads that one item or comment and writes one record to the app’s own storage. One hourly job checks whether any scheduled report is due, runs the ones that are, and does nothing when none is. Other reports load in the browser of the person looking at them: when they press Run, open a share link, or view a dashboard gadget. A load can be cancelled at any time.
Does my data leave Atlassian? The app never sends it outside Atlassian. The app declares no egress permission, and Forge blocks external calls at the platform level. The vendor has no endpoint that could receive your data and no access to your storage. If people use the app’s Rovo agent, Atlassian Rovo writes its answers under Atlassian’s own terms, using Atlassian’s sub-processors. See Permissions and security.
What happens when the trial ends? Each work item’s own field changes and comment history stay free, in the History+ tab and the panel. The explorer’s reports, exports, revert, the Deleted items tab and restore, the Reports tab, the Security scanner, the Viewed by tab, the dashboard gadgets and creating scheduled reports need an active subscription. Nothing already stored is deleted: saved views, snapshots, comment versions and view records are all kept, and deleted-item, comment and viewer capture keep running, so a lapse does not create a gap in your audit trail. Subscribe again and everything is there.
Is there a free tier? The plan is free for up to 10 users, on the same terms. There is one plan and every feature is in it, including bulk revert, the scanner, advanced reports and the gadgets.
Do I need a separate licence for each site? Licensing follows Atlassian’s Marketplace model, which is per site. Install it on each site you want it on.
What is an access code? A signed code issued to reviewers and pilot teams so they can use the full app without a subscription, for one site and a fixed period. A Jira administrator redeems it on the admin page. It is not a purchase and creates no ongoing entitlement. See Admin settings.
Why does UndoTrail ask me to “Allow access” the first time? UndoTrail reads Jira as you, through Jira’s own API, so it can only show you what your Jira permissions allow. Jira asks each person to agree to that once, the first time they open the app after it’s installed. Press Allow access and accept; it won’t ask again for that install.
Deleted work items
Can I get back a work item that was deleted before I installed the app? No. There is no source to recover it from. Capture starts at install, and the app has nothing from before that moment. This is true of every app of this kind on Jira Cloud.
Why does a restored work item get a new key? Jira assigns keys and does not allow a deleted key to be reused. The app creates a fresh work item and adds a line to its description saying which key it was restored from and when that key was deleted. Anything that pointed at the old key, such as a bookmark, a commit message or a link from another work item, still points at nothing.
Can it restore comments or attachments? Comments, yes, as copies; attachments, no. If comment history was on, the app kept each comment’s text, and the restore preview offers to re-post them on the new work item, up to 40 per restore. Jira cannot post a comment as someone else or with a past date, so each one is posted by the person restoring, now, with a line naming the original author and date. Attachment files are deleted with the work item and cannot be recreated. Links, logged work, the created date and the change history are not restored either. The restore preview lists every one of these before you commit. See Deleted items.
What does “partial” mean on a deleted item? The work item was never created or edited while the app was installed, so no full snapshot was ever taken, and only the delete event’s fields are available. That event doesn’t include the security level, so a partial item is shown only to Jira administrators. A Jira administrator can run a backfill to take a full snapshot of every work item the app can read, custom fields included, which prevents this for future deletions. A work item at a security level that leaves out the app can’t be snapshotted, so if it is deleted it may not be captured at all, or only partly.
How long do I have to restore something? The retention setting, 90 days by default, adjustable from 1 to 365 days by a Jira administrator. After that the record is removed automatically.
Can two people restore the same item twice? No. Once an item is restored, its row shows Restored as with the new key, and a second restore is refused. A restore that is still running blocks another attempt for two minutes, so two browser tabs cannot create two copies.
Reports and exports
How big an export can I do? The export writers have been tested at 250,000 rows. A report loads up to 25,000 work items (you choose the limit in the scope panel) and up to 150,000 changes. Excel splits at 1,048,575 rows per sheet and continues on the next one, and cell values over 32,767 characters are truncated with a marker, both because those are Excel’s limits. For very large jobs, split by month or project and prefer CSV.
Why does my export show two date columns? Every row carries the instant in UTC and the same instant in your own time zone, with the UTC offset in its own column. Compare on UTC when people in different countries share a file. See Exports.
Can I export to PDF? Yes. Export PDF on the Changes tab and on a work item’s history writes a readable extract of the change history, and the Reports tab exports time in value, change counts or status over time as PDF too. Above 5,000 rows the app asks you to confirm, because CSV or Excel suits that size better. See Exports.
Can I prove an export has not been changed? Yes. Tick Sign it before exporting from the Changes tab and give a reason. The app records the file’s SHA-256 fingerprint, who exported it, when and why. Later, anyone with the admin page can check the file under Verify a signed export. It is a tamper-evidence record, not an electronic signature under 21 CFR Part 11. See Signed exports.
Why does my report say it stopped at 5,000 work items? Your JQL matched more work items than the Max work items setting. Raise the limit, or narrow the query. The report always tells you when the load was truncated, so a partial report is never mistaken for a complete one.
Why is a work item missing history I can see in Jira? Most often the date range. The report filters changes to the range you chose, and a work item updated recently can have older changes outside it. Widen the range, or clear the filters. The empty-state message names the filter that removed everything.
Can I report on comment text? No. Jira’s change history does not include comment bodies, so they cannot appear in a report. Comment edits and deletions are captured separately and shown on the work item. See Tracking.
Permissions and compliance
Does it work with Jira’s issue security levels? Yes. Reports read history as you through Jira’s own API, so an item hidden from you by a security level never appears. The Deleted items list checks the same thing: a deleted item that was in a security level shows only to Jira administrators and to people Jira confirms are in that level, and only they can preview or restore it. The app compares Jira’s count of work items at that level, as you and as the app. If Jira can’t confirm it, those items stay hidden, and the list shows a short note, naming nothing, that something was left out. A partial deleted item, whose level isn’t known, shows only to Jira administrators.
How do I give an auditor read-only access? Give them a Jira account with Browse projects on the projects in scope and nothing more, then, on the app’s admin page, leave History reports and Export open to their group and set Revert and bulk revert and Deleted work items to Jira admins only, or to groups that exclude them. They can then run reports and export evidence, and cannot change anything. Reverting and restoring would also require Jira edit and create permissions, which a browse-only account does not have.
Who can use the security scanner? Jira administrators, until you choose otherwise. It starts as Jira admins only because it gathers every secret a person can see into one list. On the admin page, set it to a security team’s group, or to everyone. Whoever uses it still only scans work items Jira lets them see.
Can someone see history for a project they cannot access? No. Every read happens under that person’s own Jira session.
Who can change the settings? Only Jira administrators. The check runs on the server for every call, not just in the interface.
How do I delete the comment and viewer history the app holds? A Jira administrator can use Erase captured tracking data on the admin page, for one project or the whole site. See Tracking.
Does reverting hide who made the original change? No. Jira keeps both the original change and the revert, and the revert is recorded under the name of the person who ran it.
Compatibility
Does it track workflows as well as work items? It follows every work item through your workflow: each status change, who made it and when, alongside every other field change, comment edit and deletion. That is what answers “who moved this to Done?” and drives the time in status and status over time reports. See Reports and the explorer. Changes to the workflow’s own design, and to other Jira settings such as schemes and screens, are recorded by Jira’s built-in audit log under Settings → System → Audit log, so the two together give you the full picture.
Does it work with team-managed projects? Yes. Team-managed and company-managed projects both expose the standard change history, and both are covered by capture and restore. Restore depends on the target project’s create screen, which differs between the two types, and the preview shows exactly which fields fit.
Does it work with Jira Service Management? Yes. The app uses the standard Jira work item and changelog APIs, which Service Management projects use as well, so reports, capture and restore run the same way. We have checked deletion capture, including subtasks, on a Service Management project. Two known points: the request type field is set by the portal, so restore skips it, and SLA data is never stored, because it cannot be written back.
Does it work with Jira Product Discovery? Yes. An idea’s changes appear in reports like any other work item’s, and a deleted idea is captured and restored with its description, labels, priority, status, comments and the fields people set, such as Impact and Effort. The values Product Discovery works out for itself (Insights, Comments and Linked items counts, Delivery progress and status, Score) are left for it to work out again on the restored idea. Open an idea’s history from Apps → UndoTrail: Product Discovery’s own idea screen doesn’t show app tabs, so the History+ tab isn’t there.
Does it work with Confluence? No. This is a Jira app.
Does it work with Rovo? Yes. The app adds an agent called UndoTrail Assistant to Rovo, which answers questions about change history, deleted work items and leaked secrets, with a link to the full report. See Rovo agent.
Does it work on Jira Data Center or Server? No. It is a Forge app for Jira Cloud.
Does it support data residency? The app’s own data lives in Forge storage in the region Atlassian assigns to your site, under Atlassian’s data-residency programme.
Day to day
How do I find who changed a field last month, across ten projects? Open the explorer, select the projects, set Dates to Last month, press Run, then pick the field in Field and the person in Changed by. Export the result.
How do I find every move into Done?
Add a transition filter of → Done and leave the field on Status, or on Any field if you want the same value in any field. See Reports and the explorer.
How do I undo an automation rule that edited 300 work items? Run a report over the period, tick Bulk changes only, select the rows, and press Revert n selected. The dry run shows what can be reverted, what changed again since, and what cannot be written back. See Revert.
Why does the app show changes with no author? Jira records some changes without an author, typically automation, anonymous actions or a deleted account. They appear as Anonymous / automation and can be filtered like any other author.
What does the BULK badge mean? One person or automation changed the same field on at least 5 different work items within 10 seconds.
Is there a REST API? No. The app has no public endpoint, which keeps everything it does inside your Atlassian site and keeps it eligible for Runs on Atlassian. Reports run in the browser under the viewer’s own permissions; scheduled reports are delivered to a Jira work item. If an API matters to you, tell us what you would use it for.
Can I schedule a report by email? Yes. Choose a work item for the report to land on and add the people who want it as watchers. Each run attaches the file and posts a comment summarising it, and Jira emails everyone watching, the same way it tells them about any other comment, with the summary and a link to the work item that holds the file. The email comes from Jira itself, because the app sends nothing outside Atlassian, by design. Set one up under Settings → Marketplace apps → UndoTrail → Scheduled reports. Creating one needs an active subscription.
One thing to know: a scheduled run has no person behind it, so it is produced by the app and limited to the projects its creator could see when the schedule was saved. The delivered comment says so. See Scheduled reports.
How do I get support? Email help@astrospray.com or use the contact form and we are happy to help. Troubleshooting covers the problems people hit most often, if you would rather try first.