A serverless approach for continuous forensic artifact collection on Amazon RDS and Aurora databases is demonstrated, capturing point-in-time snapshots of database internals on a cadence and storing them in Amazon S3. This creates a time-series record for AI tools to analyze in seconds, turning hours of manual investigation into an instant conversation. The solution complements Amazon CloudWatch Database Insights by capturing a historical record with more complete information, such as full database process lists, engine statistics, and query lock lists. It achieves two objectives: capturing telemetry history by automatically collecting point-in-time forensic snapshots (active sessions, lock waits, wait statistics, query performance) on a recurring schedule and storing them as structured JSON artifacts in S3; and analyzing with AI agents using AI coding assistants (Kiro or similar) with S3 and CloudWatch access to correlate artifacts across time windows for root cause analysis. The concept involves running diagnostic SQL scripts on the target database instance on a recurring cadence and storing the output in S3 as timestamped artifacts.