Agent context
Zigpoll models
4 SQL models that turn the raw Zigpoll 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
- 5Raw tables
- 5Staging
- 5Core
- 4Analytics
Staging cleans each raw table once, core holds the logic, and analytics is the table dashboards and agents read.
- Answer detailEvery answer, in long format, with selected options unnested.grain: zigpoll_account x response x selected option (x matrix row) · 4 raw tables
- Attribution by answerOrders and revenue per answer option, for post-purchase questions such as "How did you hear about us?".grain: month x zigpoll_account x poll x slide x answer x currency · 4 raw tables
- Response volumeAnswers and respondents per poll per day.grain: day x zigpoll_account x poll · 2 raw tables
- Score summaryAverage score per rating question per month, with NPS where the scale is 0 to 10.grain: month x zigpoll_account x poll x slide · 5 raw tables
- 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 Zigpoll's own reports first. Each model's notes say what the synced data can and can't support.
Connect Zigpoll.
Your agents start with context.
Sync Zigpoll to your warehouse, deploy these models, and query them from Claude, Cursor or ChatGPT over MCP. Free for 14 days, no credit card.