Privacy Policy

What Sift Board is

Sift Board is an Atlassian Forge app that puts a table on a Confluence Cloud page and fills it from sources you configure: an inbound webhook, a scheduled connector to a third-party service, a Jira query, a CSV, or the Sift Agent running on your own machine. It runs on Atlassian's infrastructure (“Runs on Atlassian”).

Data does leave Atlassian when you configure it to

This is the single most important thing on this page, so it is stated plainly and first. Unlike its sibling app Sift, Sift Board is not a zero-egress app. It cannot be: fetching your build results from Jenkins means calling Jenkins. Concretely:

• Scheduled connectors make outbound calls. When you set up a connector for GitHub, Jenkins, GitLab, Azure DevOps, CircleCI, Bitbucket, SonarQube, PagerDuty, Datadog, Grafana, ServiceNow, TestRail, Linear, Google Sheets, or any other JSON API, Sift Board calls that service's API on the schedule you choose, from Atlassian's infrastructure, using the credential you saved. The endpoint sees that request and whatever its own logs record about it, under that service's privacy policy, not this one.
• Inbound webhooks are incoming only. A webhook view gives you a web address and a token. Any tool you give them to can post rows to that view. Sift Board makes no outbound call for a webhook view, but anyone holding the address and token can write to it, so treat the token as a secret.
• Jira queries stay inside Atlassian. A Jira view runs your JQL through Atlassian's own API and never reaches a third party.
• The Sift Agent pushes out, never in. The Agent is a separate program you run inside your own network. It connects to your database or file, then posts rows to a webhook view. Nothing dials in to your systems, and your database credentials stay in your environment and are never sent to this app.
• Nothing is sent to the developer. Board data, connector settings and credentials are never transmitted to me or to any analytics, telemetry or error-reporting service. There are none in the app.

Data we process

• Board rows. The rows shown on a board, whatever their source, are stored in Atlassian's Forge storage inside your own Atlassian site, along with the per-cell timestamps and notes that make stale values fade. They are deleted when the board or the app is removed.
• Board and view settings. Column names, filter and dropdown columns, colour rules, view names, schedules, and connector configuration such as a repository name, a project key or a JQL string. Stored in the same Forge storage.
• Third-party API tokens. The credentials you enter for connectors are stored encrypted in Forge storage inside your site. They are write-only: once saved, a token is never displayed again, to you or to me. You can replace or delete one at any time. They are used for one purpose only, calling the service you configured them for.
• Webhook tokens. Generated by the app so a tool can post rows to a view. Rotate or revoke them at any time.
• Identity. Sift Board receives your Atlassian account ID from the Forge platform, to remember each viewer's last filter selection and to ask Confluence whether you may edit the board. Names and email addresses are not read.

Data we do not collect

No analytics, no telemetry, no cookies, no error-reporting service, no marketing. The app makes no network call other than the connector calls you configure and the Atlassian APIs it runs on.

Permissions

• storage:app: the board rows, settings and encrypted credentials described above, held inside your site.
• read:confluence-content.permission: a permission check to decide whether you may edit a board.
• Jira read access: used only when you add a Jira view, to run your JQL and read the fields you asked for.
• External (egress) permissions: the domains a connector needs to reach. Confluence shows you these when you install the app, and a connector cannot call anywhere the app has not declared.

Your responsibility for what you connect

You choose which services a board talks to and which credential it uses. Give each connector the least access it needs, prefer read-only tokens, and check that pulling that data into a Confluence page is allowed under your own policies and under the terms of the service you are connecting.

Retention and deletion

Board data, settings and credentials live for as long as the app is installed on your site. Deleting a view deletes its rows and its credential. Uninstalling the app removes everything it stored. Site administrators may also contact us to request deletion.

Sub-processors

Atlassian, for hosting and Forge storage. No others. The third-party services you connect are not sub-processors of ours: you configure them directly, and your relationship with them is governed by their terms.

Your rights

Questions or requests under GDPR or similar laws: contact us at the address below. We answer within 30 days.

Changes

We will post changes to this page and update the effective date. If a change widens where data travels, it will be called out at the top of this page.

Contact

Aravind Arumugam, Chennai, India · aravind@arumugam.co.uk