Privacy Policy

AutoPrioritizer AI for Jira Cloud

Last updated: April 22, 2026

Overview

AutoPrioritizer AI is an Atlassian Forge app that uses artificial intelligence to help teams prioritize Jira tickets. This privacy policy explains what data the app accesses, where it goes, and how it is stored.

We believe in transparency. This policy is written in plain English so you know exactly what is happening with your data.

What Data Does the App Access?

When you prioritize a ticket (individually or in batch), the app reads the following fields from your Jira issues:

• Summary (ticket title)
• Description (ticket body text)
• Issue type (e.g., Bug, Story, Task)
• Status (e.g., To Do, In Progress)
• Priority (current priority level)
• Labels
• Components
• Comments (the three most recent comments only)

This is the minimum set of information the AI needs to make an informed prioritization decision. No other Jira data is accessed during prioritization.

Where Is Data Sent?

Ticket data listed above is sent to Anthropic's Claude API (api.anthropic.com) for analysis. The AI evaluates each ticket against your team's configured criteria and returns a recommended priority level with a short explanation.

Anthropic's data handling

Anthropic does not use API inputs or outputs to train its models. Per Anthropic's API Terms of Service, data submitted through the API is not used for model training or improvement. Your ticket data is processed for the sole purpose of generating a prioritization response and is subject to Anthropic's data retention policies for API usage (typically 30 days for trust and safety purposes). For full details, refer to Anthropic's API Terms and Privacy Policy.

What is sent to Anthropic

Each API request contains:

• A system prompt describing your team's prioritization criteria, weights, custom rules, and team context (all configured by you)
• The ticket fields listed above, stripped of HTML and truncated (descriptions to 2,000 characters, comments to 600 characters each)

No user account information, email addresses, or Jira credentials are included in API requests.

What Data Is Stored?

AutoPrioritizer AI runs entirely on Atlassian Forge infrastructure. All persistent data is stored in Forge Key-Value Store (KVS), which lives within Atlassian's infrastructure and is scoped to your Jira instance.

Stored in Forge KVS (per project):

• Prioritization profile -- your team's criteria weights, team context description, priority level definitions, and custom rules
• Quick filters -- saved JQL filter shortcuts for batch prioritization
• Usage statistics -- total tickets prioritized, monthly usage counts, session history (timestamps and counts), and monthly active user counts for quota tracking

NOT stored anywhere:

• Ticket content (summaries, descriptions, comments) is never stored by the app. It is read from Jira at the time of prioritization, sent to the Claude API, and discarded.
• AI responses (priority recommendations and reasoning) are returned to the browser and used to update the Jira issue. They are not persisted by the app.
• No data is stored on any server outside of Atlassian's Forge infrastructure. There is no external database, no analytics service, and no third-party storage.

Data Retention

• Forge KVS data (profiles, filters, usage stats) persists as long as the app is installed on your Jira instance. Uninstalling the app removes this data per Atlassian's Forge data retention policies.
• Anthropic API data is subject to Anthropic's retention policies. Per their terms, API inputs may be retained for up to 30 days for trust and safety purposes, after which they are deleted.
• The app itself retains zero ticket data. Nothing is cached, logged, or saved between requests.

Permissions

The app requests the following Atlassian scopes:

Scope	Purpose
read:jira-work	Read issue fields (summary, description, labels, etc.) to send to the AI for analysis
write:jira-work	Update the priority field on issues after you approve the AI's recommendation
storage:app	Store prioritization profiles, quick filters, and usage statistics in Forge KVS

The app also has permission to make outbound requests to api.anthropic.com for AI analysis.

How write access is used

The app only writes to the priority field on Jira issues. It does not modify summaries, descriptions, comments, or any other fields. Priority changes are always initiated by user action -- either clicking "Apply" on a single ticket, "Apply All" on a batch, or through the auto-prioritize trigger (which the user must explicitly enable per project).

API Key

The app requires an Anthropic API key to function. This key is provided by the Jira administrator and stored as a Forge environment variable (set via forge variables). The key is:

• Stored securely by Atlassian's Forge platform as an encrypted environment variable
• Never exposed to the frontend or to end users
• Used solely to authenticate requests to Anthropic's Claude API

You are responsible for managing your own API key and any associated costs on your Anthropic account.

Third-Party Services

The only third-party service the app communicates with is:

Service	Purpose	Data Sent
Anthropic Claude API (api.anthropic.com)	AI-powered ticket analysis	Ticket fields (summary, description, labels, components, comments, type, status, priority) and your prioritization criteria

No analytics, tracking, advertising, or monitoring services are used.

Children's Privacy

This app is designed for use by Jira Cloud teams in a professional context. It is not intended for use by children under 13.

Changes to This Policy

We may update this policy from time to time. Material changes will be communicated through the Atlassian Marketplace listing. The "Last updated" date at the top of this page reflects the most recent revision.

Contact

If you have questions about this privacy policy or how your data is handled, contact:

AutoPrioritizer
Email: support@autoprioritizer.app