HR, Finance, and Engineering ask "how many employees do we have?" and get three different numbers — from the exact same underlying data. Workforce Astra encodes the definition once, as a governed Snowflake Semantic View, so every team gets the identical, SQL-transparent answer.
Ask "how many employees do we have?" two ways. Same 150-row dataset, same moment in time — the numbers genuinely disagree, because one query is governed and one isn't.
These are the 6 AI_VERIFIED_QUERIES registered on employee_360.
Click one — the answer and the SQL that produced it come from the real dataset, live.
Comp-ratio below 0.85 and a rating of 4 or above — underpaid high performers, the real retention risk. Click a name for the cited review evidence, then simulate the action a CoCo agent would take via the mock MCP server in this repo.
3 tables, 3 metrics, 6 verified queries. Every question resolves through employee_360, never raw tables.
Synthetic data generation packaged as $workforce-astra-data-gen — generate, verify, stage, load.
Two tools — Slack manager alert, Workday comp-adjustment draft — protocol-tested with the real MCP client.