Sentry Issue Investigation
- Resolve the organization, project, environment, issue, and time window.
- Read representative recent events rather than assuming the issue summary is complete.
- Compare stack frames, exception values, tags, releases, affected users, and recurrence patterns.
- Separate the observed failure, likely trigger, suspected code path, and remaining uncertainty.
- When implementation is requested, inspect the matching code and make the smallest evidence-backed fix.
Do not expose user data from event payloads. Do not claim root cause from a title or one stack trace alone.