Anomaly Sweep
Weekly check on top metrics, reports only real moves.
Data Adapted from use case · Added Aug 22, 2026
Based on 231 ways to use an agent by @dhruvamin
Run Anomaly Sweep yourself. Make a Skydive agent (name it, give it a purpose), then send it the prompt below as your first message. It asks for what it needs and sets itself up. Free to copy, adapt, and edit after.
THE SETUP PROMPT
Set up a new agent named Anomaly Sweep to watch our top metrics without generating noise. Walk me through connecting Snowflake, Looker and Slack, then: pull the top 30 tracked metrics, compare each against its historical range and seasonality, and filter out normal variance so only statistically real movements remain. Post a Slack report listing each real anomaly with the metric, the size of the move, and the likely driver. Learn which metrics are noisy versus stable and their normal seasonal pattern from a year of history, then save it and run weekly.
CONNECT FIRST
Snowflake
Looker
Slack TALKS TO YOU IN
SlackEmailiMessage
Every Skydive agent ships with these channels. Connect one in your workspace and it gets its own inbox or number. Not agent-specific.
SKILLS INSIDE
anomaly-detectionmetric-triage
Reusable capabilities the prompt sets up. Browse the open skills ecosystem at skills.sh.
See something off with Anomaly Sweep? Suggest an edit. Every agent is one markdown file in the repo.