Security highlights
- No external network calls (no outbound HTTP requests).
- No data egress (all processing stays within Atlassian’s cloud).
- Minimal permissions: only the 3 scopes required to function.
- No secrets stored (no API keys, passwords, tokens, or credentials).
Forge platform security
WatcherSync is a Forge-native application, meaning it runs entirely within Atlassian's managed infrastructure. There is no self-hosted server component.
- Sandboxed execution: code runs in an isolated, managed runtime environment controlled by Atlassian.
- Managed hosting: Atlassian handles infrastructure, scaling, and availability.
- Built-in app identity: uses Forge authentication (
asApp()) without stored credentials. - Content Security Policy: Forge enforces strict CSP headers to reduce XSS and exfiltration risk.
See Atlassian’s Forge security documentation: https://developer.atlassian.com/platform/forge/security/
Network security
WatcherSync makes zero external network calls. The only communication is between Forge runtime and Jira’s internal APIs within Atlassian’s infrastructure.
Data storage & encryption
All persistent data is stored in Atlassian Forge Storage (encrypted at rest and in transit, tenant-isolated). Stored data is operational and non-sensitive: project enable/disable toggles, aggregate counters, timestamps, and idempotency markers.
No personal data, user identifiers, issue content, or credentials are written to storage.
Permissions (scopes)
WatcherSync requests the minimum scopes required:
read:jira-work— read watcher lists and issue context to determine what to copy.write:jira-work— add watchers to subtask issues.storage:app— store per-project configuration and aggregate counters in Forge Storage.
The app does not request scopes for managing users, accessing admin settings, reading/writing issue content beyond watcher operations, or making external API calls.
Incident response
In the unlikely event of a security concern related to WatcherSync, contact [email protected]. We’ll investigate and respond within 48 hours.
Contact
For security-related questions or to report a vulnerability: [email protected]