examplepay — effective contract = spec + traffic overlay (upstream "examplepay")
overlay version: 3 (sandboxes serve the latest; every response carries x-pikopod-contract-version)
OBSERVED additions (traffic-admitted; the sandbox renders these):
v1 field GET /charges/{id} 2xx refund_reason (string, presence 0.41)
v2 value GET /charges/{id} 2xx status += "disputed"
contradictions (spec vs traffic — both sides kept, never erased):
GET /charges/{id} 2xx amount: spec says number, traffic says string (99% of clean samples) — winner: traffic
pinned scenarios (immovable — each runs at its pin-time contract):
drift-385153d1776c pinned at v1 — 2 version(s) behind
observed state machine — examplepay
GET /charges/{id} · status (412 transitions over 9 days)
pending → succeeded 380
pending → failed 31
never observed: failed → pending, succeeded → pending
