Nexus privacy policy

Last updated 10 October 2026

This policy explains what information Nexus (the Jira app) and the website nexus.resync.nz handle, where it is kept, and who can see it. Nexus is made by Resync Consulting Limited, a New Zealand company ("Resync", "we", "us").

The short version

Nexus runs on Atlassian Forge, Atlassian's own cloud platform. Resync operates no servers of its own, and the information you put into Nexus is stored in your Jira site and in Forge storage inside Atlassian's cloud. Resync does not receive it.
Nexus sends nothing outside Atlassian unless your Jira administrator connects an AI account or a migration source, or a project administrator connects a CI service (and someone presses Run in CI).
AI: Nexus sends requirement content to the AI provider your Jira administrator connects, with your own account and key, only when someone runs an AI action. Resync never sees that content.
Migrating from another test tool: if your administrator connects Xray, Zephyr Scale, Zephyr Squad or AIO Tests, Nexus reads test cases from that service into Jira. No Jira data is sent to it.
CI services: if a project administrator connects your own GitHub Actions, GitLab.com, Azure DevOps or Bitbucket Pipelines account, Nexus sends that service a few run details (see "Starting a pipeline from Nexus") when someone presses Run in CI.
Read API: a project administrator can create an API key that lets whoever holds it read that project's test records. It is off unless an administrator chooses it.
Nexus stores Atlassian account IDs for audit trails, but never names, email addresses or avatars.
We do not sell personal information, and we do not use your data to train AI models.

Who is responsible for what

For the information your organisation puts into Nexus (requirements, test cases, runs, test data, environments, releases and so on), your organisation is the data controller and Resync is a data processor acting on your organisation's instructions. If you are a user of Nexus at a company that installed it, your Jira administrator is usually the right first contact for questions about that data. Our Data Processing Agreement sets out the terms on which Resync processes that data for your organisation.

Resync is the controller only for the small amount of information we handle directly: emails you send us, and the account and licensing details Atlassian shares with app vendors (see below).

What Nexus stores, and where

Everything Nexus stores stays inside Atlassian's platform:

On your Jira issues. Test Case, Test Set and Test Execution are ordinary Jira issues. Test steps, Gherkin text and similar design data are stored as properties on those issues. They are your Jira data and follow your Jira permissions.
Cold storage (off unless a Jira administrator turns it on). To save space in Forge storage, a Jira administrator can choose to move older details out of it and into issue properties on the Jira issue they belong to: the step details of finished test runs (on the Test Execution), older saved versions of a test case (on the Test Case; the newest version always stays in Forge storage) and the last requirement text excerpt Nexus keeps for re-analysis (on the requirement). Anyone who can view that issue in Jira can read these properties through Jira's REST API, which matches who can see them in Nexus. They are deleted at the same retention window as before, and turning cold storage off and choosing "Undo: bring moved details back" moves them back into Forge storage and deletes them from the issues.
In Forge storage for your site. Run results and step results, test case version history, test data sets, environment and deployment records, releases and sign-offs, report snapshots, draft AI findings and drafts awaiting review, settings, and audit records of who changed what.
Identifiers, not profiles. Where Nexus records who did something, it stores the person's Atlassian account ID only. Names and avatars are looked up from Jira when a page is shown and are not stored by Nexus.
When a person is removed. Nexus learns that a person has gone in two ways: Jira tells it when a user is deleted from the site, and once a week Nexus asks Atlassian's personal data reporting service which of the account IDs it stores belong to closed accounts (it sends only account IDs). Nexus then deletes that person's own records (their preferences, saved views, private report views and AI step suggestions) and replaces their account ID with "unknown" in settings, run history, shared views and similar working records. Assignments they held, such as a default bug assignee or a run step assignee, are cleared. Normally this finishes within the hour. Data held in Jira itself (issues, comments and mentions, assignees, and the Nexus data saved on issues) follows Jira's own handling of removed accounts, which shows them as a former user.
Audit records keep the account ID. Nexus does not remove the account ID from its audit records: the configuration change log, requirement finding review decisions, coverage exclusion history, test cycle history (who closed, reopened, locked or archived a cycle, or added or removed its test cases), release sign-offs and release status changes, deployment and environment event records, and test case change history, and the per-step result history (which tester recorded each step result and when). An audit trail has to say who approved or changed something to be useful as evidence, for example in a quality or regulatory audit, and your organisation has a legitimate interest in keeping it complete. Only the bare account ID is kept (never a name or email address), Jira shows the person as "Former user", and these records are still deleted at the end of the audit retention period your administrator sets (7 years by default).
Keys. CI API keys are stored only as salted hashes and shown once. An AI provider key your administrator connects is kept in Forge's secret storage and is never sent back to the browser. The same applies to the access token for a CI service.

Forge storage is operated by Atlassian and follows Atlassian's data residency: where your Jira site's data is pinned to a region, Nexus's Forge storage follows it. Atlassian encrypts this data in transit and at rest.

Nexus follows your Jira permissions. It checks the viewing user's own Jira permissions, including issue-level security, before showing or changing anything.

Test data. Nexus's test data sets are meant for made-up (synthetic) values. Nexus scans data sets for likely real personal information, always refuses real card numbers, and masks values in sets marked sensitive. Please don't put real customer data into test data sets.

AI processing

Nexus uses the AI account your Jira administrator connects: Anthropic, OpenAI, Google, Moonshot (Kimi) or xAI (Grok), with your own API key. A Jira administrator can turn AI off for the whole site, and a project administrator can turn it off for their project; while it is off, Nexus sends nothing to the provider for that site or project. When someone runs an AI action, Nexus sends that provider only what the action needs, such as the summary, description, acceptance criteria and approved findings of the one requirement being analysed or generated from, or the steps of the test case being worked on. Issue comments are sent only when the person running the action chooses to include them. Rows of test data sets marked masked or sensitive are never sent to the provider as examples, and a value from a set marked sensitive is replaced with a placeholder such as [SENSITIVE_DATA_1] wherever it appears in a test case sent to the provider, whether or not redaction (below) is on. That placeholder is not replaced with the real value in AI drafts.

If the person running the action leaves Use attachments on, Nexus also sends the text of the requirement's own Jira attachments they choose (PDF, Word, text, Markdown or CSV files) and, for providers that accept images, the chosen PNG or JPG images. These are the same kind of content as the requirement itself and go to the same provider. Nexus reads them from Jira for that one request and stores none of their content.

That content goes directly from Atlassian's Forge platform to the provider. Resync runs no server in between and never receives it. How the provider processes, stores and uses it is governed by your organisation's own agreement with that provider, including which region it is processed in. Nexus does not offer region-pinned AI processing.

Personal data redaction (optional). A Jira administrator can turn on redaction for the whole site (a project administrator can turn it on for their own project when the site setting is off). Nexus then replaces the kinds of data the administrator selects — such as email addresses, phone numbers, payment card and bank account numbers, national and tax IDs, IP addresses, API keys and tokens, Jira user mentions and dates of birth, plus the administrator's own words and patterns — with placeholders such as [EMAIL_1] before the text is sent, and puts the real values back into the drafts it shows you unless the administrator chooses to keep the placeholders. It also masks values in test data columns whose heading names personal data (such as name, surname, email, phone, address, postcode or date of birth). Detection is rule-based, not AI, and it is not complete: it does not reliably catch people's names or postal addresses written in ordinary text, and it cannot scan images. Treat it as a safeguard that reduces what is sent, not a guarantee that no personal data reaches the provider. The list of real values exists only for the one request and is never stored. Because images cannot be scanned, images from attachments are not sent to the provider while redaction is on, unless an administrator turns on "Also send attachment images (not scanned)"; the AI page tells you when images were held back. The same settings apply to what Nexus returns to Atlassian Rovo when you ask Rovo about your tests.

Even with redaction off, Nexus always replaces private keys, JSON Web Tokens, recognised API keys and access tokens, and passwords written into web addresses before anything is sent to the provider or returned to Rovo. With redaction off, other personal data in the content described above is sent as written.

Nothing is sent when nobody runs an AI action. Until a Jira administrator connects an AI account, Nexus uses no AI and sends nothing to any AI provider.

Every AI finding and every AI-drafted test is a draft: nothing is written to your Jira issues until a person accepts it. AI output is a suggestion, not an authoritative or binding quality decision.

Migrating from another test tool

If a Nexus administrator connects Xray Cloud, Zephyr Scale Cloud, Zephyr Squad Cloud or AIO Tests Cloud to migrate test cases, Nexus uses the API credentials they enter to read test cases from that service into your Jira site. The credentials are stored encrypted in the app's Forge storage on your site, used only for that migration, and deleted when it finishes (by default) or after 30 days. No Jira data is sent to those services.

What is read is the test content itself: titles, descriptions, preconditions, steps, folder names, labels, priority, status and custom field values, written into your Jira site as Nexus test cases. What is sent to the other service is only the credentials and which project to read. Vendor user accounts are not copied.

Starting a pipeline from Nexus

If a project administrator connects your organisation's own GitHub Actions, GitLab.com, Azure DevOps Pipelines or Bitbucket Pipelines account, a user can press Run in CI on a test cycle. Nexus then sends that service only the values the administrator ticked: the project key, the cycle's key and label, the environment and build the user typed, and the address Nexus receives results on. No Jira issue content and no Nexus API key is sent. Nothing is sent until an administrator has connected a service and a user presses Run in CI.

The access token the administrator enters is kept in Forge's secret storage, is never shown again or written to logs, is left out of backups, and is deleted when the connection is removed. The CI service is your organisation's own, under your organisation's agreement with it, and its own terms govern what it does with the values it receives (some services show them on the pipeline's page). Nexus records who started each run, as an Atlassian account ID, for 30 days.

Reading test data through the API

A project administrator can create an API key with the Read test data permission. It is never added to existing keys or to new keys by default. Whoever holds the key can read that one project's test cases, cycles, runs, step results and coverage. Issues with an issue security level are never returned, values in test data sets marked sensitive are always masked, and people appear only as Atlassian account IDs. Giving the key to someone is your organisation's instruction to share those records with them. Nexus records one entry per key per day in the project's audit log.

How long information is kept

While Nexus is installed, the app's own working history (run detail, report snapshots and similar records) is removed after 180 days by default. Your Jira administrator can set this per project, from 30 to 3,650 days. Findings still waiting for review are kept until someone reviews them.
Audit records are kept for 7 years by default, adjustable per project from 1 to 10 years, because they serve as audit and release evidence: deployments, environment events, releases and go/no-go sign-offs, test case change history, reviewed AI findings (who accepted, edited or rejected them), the administrator change log, and the history of requirements marked "Not testable". To save space, test case history entries older than a year keep who, when and what changed but drop their copy of the test steps (the newest entry for each test case is always kept whole).
Audit records are not complete evidence for the whole audit window. The 7-year window keeps the record of what happened (who, when, which decision). Detailed evidence is kept for shorter, fixed periods: per-step run detail (step results, comments, per-step evidence) is summarised after 90 days by default (30 to 3,650 days per project; the counts, failed steps with their first comment and linked defects, and run duration stay), except runs in cycles that a release was last signed off Go against, which keep full detail while the release exists; a test case's design copy on history entries older than a year is removed (the newest is always kept); up to 50 named versions per test case; and requirement text at a Go sign-off for the first 500 requirements, for the newest five Go sign-offs of each release (the latest always kept). Nexus shows "Evidence is incomplete" on a release's Readiness tab when any of these applies, and a project can turn on strict sign-off to refuse a Go whose evidence would be incomplete.
Backups. Atlassian's site backup does not include app data, so a Jira administrator can export everything Nexus stores (Nexus Setup → Backup & Restore) and restore it later. A backup file never contains secrets such as API keys or your AI key, and it stays wherever your administrator saves it.
Archive & purge. A Jira administrator can also move a project's old, finished test run detail (step results, comments, evidence links, run timings) out of Nexus earlier than the retention window: Nexus saves it to a file the administrator downloads, checks that file, and only then deletes those records from Nexus storage (Nexus Setup → Archive & purge). The file stays wherever the administrator saves it and can be restored. Audit records are never removed this way, and the Test Executions in Jira keep their results.
Your Jira issues are never deleted by Nexus. Test cases, sets, executions, their links and properties are your Jira data and stay in Jira if you uninstall the app. This includes anything cold storage moved into issue properties: to remove it before uninstalling, turn cold storage off and choose "Undo: bring moved details back" in Nexus Setup → App Storage.
After uninstall, Atlassian retains the app's Forge storage for a limited period under its platform retention policy and then deletes it permanently.

What Resync can see

Resync does not have access to the content of your Nexus data. We can see:

Operational logs from the Forge platform, which Nexus writes with status codes, counts, error types and opaque reference codes only, never issue text, test content, findings, issue or project keys, or account IDs. When two log lines concern the same issue, Nexus records a one-way reference code in place of the key, so the logs alone do not reveal which issue it was.
Installation and licence information that Atlassian shares with app vendors, such as your site address, edition, licence status and the technical and billing contacts you gave Atlassian. We use it to provide and license the app and to contact you about it.
Anything you send us yourself, such as a support email with a screenshot. If you send us content from your Jira site so we can help, we use it only to answer you and delete it when it is no longer needed.

Support and this website

When you email nexus-support@resync.co.nz or nexus@resync.co.nz, we keep the email and your contact details to answer you and to keep a record of the support we gave.

The website nexus.resync.nz has no sign-in and no advertising. It offers optional Google Analytics measurement so we can understand which public pages are useful. Analytics is not loaded unless you choose Allow analytics. Advertising storage, advertising user data, advertising personalisation and Google signals remain disabled. If you initially decline, no Google Analytics tag is loaded. If you later withdraw consent, measurement is disabled and Analytics cookies are removed; the already-downloaded script may remain in the page until you leave or reload it. Your choice is saved in your browser's local storage and can be changed with the Analytics choices control on the site.

When enabled, Google Analytics may process information such as the page visited, approximate location, browser, device and interaction timing under Google's terms. Do not enter personal or confidential information into public website URLs. Our web host also keeps standard server logs (such as IP address, browser and pages requested) for security and operation. The site's font is served by Google Fonts, so your browser requests it from Google.

Who else is involved

Atlassian hosts Nexus and its storage as the Forge platform provider, under Atlassian's own privacy policy and data processing terms.
The AI provider your organisation connects (if any) processes content under your organisation's agreement with it, as described above.
A test tool your administrator connects to migrate (if any): Nexus reads from it with the credentials your administrator entered; nothing from Jira is sent to it.
A CI service your project administrator connects (if any): GitHub, GitLab.com, Azure DevOps or Bitbucket. It receives the run details described above under your organisation's own agreement with it.
Whoever holds a Read test data API key your administrator creates (if any) receives that project's test records, as described above.
Google, when you opt in to website analytics, processes the limited website-use data described above and serves the site's fonts under Google's terms.
Our website hosting provider processes standard server logs to operate and secure the public website.

We do not share personal information with anyone else, except where the law requires it or as otherwise described in this notice.

International transfers

Resync is based in New Zealand. The European Commission recognises New Zealand as providing an adequate level of data protection. Data stored by Nexus stays in Atlassian's platform, and Atlassian's own transfer mechanisms apply to that hosting.

Your rights

Under the New Zealand Privacy Act 2020, and where it applies the EU and UK GDPR and similar laws, you can ask for access to personal information about you, ask for it to be corrected or deleted, and object to or restrict some uses of it.

For information inside Nexus, contact your organisation's Jira administrator first, because your organisation controls that data; we will help them respond. For information Resync holds directly, email nexus@resync.co.nz.

If you are not happy with our response, you can complain to the New Zealand Office of the Privacy Commissioner (privacy.org.nz) or to the data protection authority where you live.

Security

Nexus is built on Atlassian's security controls (encryption in transit and at rest, Atlassian's access controls) and declares the Jira and Confluence permissions (scopes) listed in its Marketplace listing, and always checks the acting user's own Jira permissions. It uses the manage:jira-configuration scope only to create the issue types, link types and workflow rules that Test Case, Test Set and Test Execution setup needs, and Confluence access only to publish reports as pages when an administrator connects Confluence. To report a security concern, email nexus-support@resync.co.nz with "Security" in the subject line.

Changes to this policy

We will update this page when Nexus or our practices change, and change the date at the top. Significant changes will also be noted in the app's release notes.

Contact

Resync Consulting Limited, New Zealand Privacy questions: nexus@resync.co.nz Support: nexus-support@resync.co.nz