Privacy Policy

Last updated: March 2026 · Ticket Intelligence · Team Tier v1.0

1. Overview

Ticket Intelligence ("the App") is an Atlassian Forge application that provides AI-powered incident intelligence within Jira. This privacy policy explains what data the App accesses, how it is used, and where it is sent.

The App is built on Atlassian Forge, which means all server-side code runs within Atlassian's infrastructure. The App developer does not operate independent servers that receive your data.

2. Data Accessed

Jira data

• Issue summaries, descriptions, resolution notes, labels, priority, and issue type — used to generate semantic embeddings for search
• Issue assignee display names and account IDs — stored as resolver attribution in the search index
• User email addresses — used transiently to look up Slack or Microsoft Teams user IDs for war room invitations. Email addresses are never stored.
• User active/inactive status — used to filter deactivated accounts from recommendations

Confluence data

If the Confluence connector is installed, the App accesses Confluence page titles and content summaries to surface related runbooks in the panel. Page content is vectorised and stored in the search index.

3. External Services

No data is sent to any external service that has not been explicitly configured by your organisation's administrator.

OpenAI API

Issue text (summary and description) is sent to the OpenAI API for vectorisation using the text-embedding-3-small model. Data is processed subject to OpenAI's API terms. Data sent via the API is not used to train models.

Pinecone

Vector embeddings and ticket metadata (issue key, summary, assignee name and ID, resolution date, resolution text, priority, issue type, labels) are stored in Pinecone for semantic search with namespace isolation per organisation.

Slack

When a war room is assembled, the following data is sent: issue key, title, URL, AI analysis text, and Slack user IDs of invited experts. Slack bot tokens are stored encrypted in Forge Storage.

Microsoft Teams / Microsoft Graph API

Similar data is sent when creating Teams war rooms. Azure AD credentials are stored encrypted and only used for authentication.

GitHub

When a P1 issue is created, issue data and AI analysis are sent to GitHub. Private keys are stored as encrypted Forge variables.

GitLab

AI analysis text is posted to GitLab issues via API. Tokens are stored encrypted in Forge Storage.

Datadog / Sentry / Linear

These services send webhook payloads to the App. The App does not send data to them.

Agent endpoint (if configured)

Running automated fixes sends issue data and analysis to the configured endpoint controlled by your organisation.

4. Data Storage

All credentials and configuration are stored in Atlassian Forge Storage, encrypted at rest. Usage data expires after 24 hours. War room metadata is temporary and deleted after completion.

5. Data Retention

Vector embeddings and metadata in Pinecone are retained indefinitely unless deletion is requested.

User attribution is retained for knowledge continuity, but deactivated users are excluded from recommendations.

6. User Rights

Jira administrators may request removal of their organisation's data at any time. Individual user attribution can also be removed upon request.

7. Security

All credentials are encrypted. Webhook signatures are verified using HMAC-SHA256 or shared secrets. The App runs entirely within Atlassian Forge infrastructure.

Scope	Purpose
read:jira-work	Reading issue data for semantic search and analysis
write:jira-work	Posting prevention comments and alert tickets
read:jira-user	Looking up user emails (transient)
search:confluence	Searching for runbooks
read:confluence-content.all	Reading Confluence content
read:confluence-space.summary	Reading space metadata
storage:app	Storing configuration and usage data

8. Contact

For privacy enquiries, data deletion requests, or security concerns:

Email: help@unobtainableverge.com