# VEP P0 — data templates
#
# Drop real GST exports here as CSV with these exact columns. The harness (run.py)
# reads them and emits KILL/KEEP/REVISE. Until files exist (or columns are empty),
# the harness reports INSUFFICIENT — it never invents numbers.

signups.csv          columns: email, intent_bucket, converted_to_foundation, date
cohort_jan2026.csv   columns: attendee, bought_ageless, upgraded_tier2
subscribers.csv      columns: sub_id, tier, practice_signal, continued, purchase_recency_days
capped_clients.csv   columns: client_id, revenue_stream, converted_to_scalable

Where to get these (GST-owned data; not available to the assistant):
  - signups.csv        : export the 190 7-Day Revival signups; tag each with the intent
                         bucket they pick in the H1 test email; mark who bought Foundation.
  - cohort_jan2026.csv : from the delivered Movement Lab cohort — who bought Ageless / Tier 2.
  - subscribers.csv    : export Tier 1/Tier 2 subscribers + any practice/view signal + churn.
  - capped_clients.csv : live-class / private-session clients; did they buy video/equipment?

Run:  python scripts/p0/run.py --data-dir scripts/p0/sample_data
      H1_THRESHOLD=0.10 python scripts/p0/run.py   # e.g. need >=10% conversion to build