Privacy and data
TuneCraft runs entirely on Atlassian's Forge platform, keeps the minimum it needs, and never keeps the texts of your flows or the names of your Jira configuration. This page says what that means.
On this page
Runs on Atlassian
TuneCraft sends nothing outside Atlassian. It has no calls to outside services, no outside scripts, fonts, images or analytics, and no outside AI service. Its data lives in Forge storage, in your site's data residency location. Its pages inherit your site's theme and each person's language.
TuneCraft does not change Jira without a person's confirmation. It checks which flows use an object before you remove it, uses Jira's own trash and archive where they exist, and keeps a record with an undo where Jira allows one.
What happens to your automation export
Your browser reads the export. The server receives summaries and, from a scan a Jira admin starts, each flow's structure with every text in its settings replaced by a placeholder made in your browser.
- Never leaves your browser: web request addresses and headers, email addresses, JQL, smart values, custom field names, values compared, comments and secrets. The browser alone puts the real values back into a file you download.
- JQL goes from your browser only to your own site's Jira, as you, to be checked and counted. It never reaches TuneCraft.
- Kept beside a flow: its id, name, trigger type, projects, creation day, failure-email setting, and its owner's and actor's account ids. A kept flow holds in clear only numbers, true or false, the export's vocabulary (an advanced schedule's cron included), Jira's own field names, project ids and a basic schedule's frequency, start and end.
- The server refuses a flow that holds anything else, and no page shows a kept flow's placeholders.
- From an audit log: only numbers and days, such as each flow's runs and steps, how its runs ended, its longest run and the day of its last run. Never the work items, people, rule names, times of day or audit log ids.
What the clean-up tools keep
From Jira's configuration, TuneCraft stores ids, counts, days, Jira's own vocabulary and keyed hashes only: never names, JQL, label text, option values, file names or work item content. The pages read names from Jira, as you, when they show them. The tool pages in this manual list what each one keeps.
People
TuneCraft stores Atlassian account ids, never email addresses or names. They appear in scans (owners, actors, who scanned), settings (who saved them), alert recipients and owner notices. TuneCraft reports the account ids it stores to Atlassian once for each reporting cycle, and erases the data of accounts that were closed. It never shows who is in a group.
Kept and removed
| Data | How long |
|---|---|
| The last 24 scans | Until a newer scan pushes them out, or you delete a scan on History |
| What flows use, for Check before you delete | The last scan only |
| Run health, usage, alerts, Limits history and Jira points a day | 400 days |
| Change records (Filters, Labels) | 400 days; the edited objects 30 days |
| A tool's readings of Jira | Until a Jira admin turns its background checks off; then removed within a day |
Maintenance removes expired data every day. When you uninstall, Forge soft-deletes the app's data. A reinstall is a new installation; Atlassian can relink the old data on request within 21 days.
Access and logs
Every call is checked on the server for the caller's access, and for the licence. When a check cannot complete, TuneCraft denies and says why. Logs carry counts and ids only, never values.
Accessibility and languages
Every visible string is available in English, French, German, Spanish, Italian, Japanese, Simplified Chinese and Hindi. TuneCraft aims for WCAG 2.2 AA: full keyboard use, visible focus, labels on every control and no meaning carried by colour alone.
Questions about the app? Write to info@plugincraft.ca.