Agent context
Mixpanel models
4 SQL models that turn the raw Mixpanel tables into the reports people ask about. Deploy them in Weld and your agent queries the definitions behind every number.
How the set is built
- 2Raw tables
- 2Staging
- 4Core
- 4Analytics
Staging cleans each raw table once, core holds the logic, and analytics is the table dashboards and agents read.
- Active users dailyDaily, weekly and monthly active users and stickiness, per day: the DAU/WAU/MAU view from Mixpanel's Insights and its "Active users" template.grain: one row per day (UTC) · 1 raw table
- Events dailyMixpanel's Insights "Total events" and "Unique users" per event, per day.grain: one row per day (UTC) x event name · 1 raw table
- Funnel conversion dailyYour saved Mixpanel funnels, step by step per entry day: how many users reached each step, the conversion from the previous step and from the start, and the time it took.grain: one row per funnel x day (users' entry day) x step · 1 raw table
- SessionsUser sessions rebuilt from raw events, using Mixpanel's default timeout rule: a session ends after 30 minutes without an event from that user.grain: one row per session (distinct_id x session start) · 1 raw table
- Weld or dbt
- The same SQL in both dialects. Sync the folder with GitHub Sync, or copy the dbt models into your project.
- Check before you rely on it
- Reconcile a month against Mixpanel's own reports first. Each model's notes say what the synced data can and can't support.
Connect Mixpanel.
Your agents start with context.
Sync Mixpanel to your warehouse, deploy these models, and query them from Claude, Cursor or ChatGPT over MCP. Free for 14 days, no credit card.