what this will do
- capture brand + tenant details to generate a .json payload per logic.* vertical
- validate against
docs/onboarding/SMART_ONBOARDING.schema.json - export payloads as
sample.<tenant>.onboarding.json(and date-stamped backups) - hand off to
scripts/onboarding/load_onboarding.py <payload.json>for seeding
initial flow will target logic.fleet (demo instance: ittd.fleet.logic.ittd.works), then expand to other logic.* verticals.
checklist (assistant script)
- brand: name, logo, colors, PDF header/footer needs
- tenant basics: locales, tax, currency, regions/territories
- roles & users: admins, dispatch, engineers, partner logins
- catalogue: job types, SLAs, pricing rules, safety/quality steps
- inventory: depots, vans, suppliers, initial stock levels
- docs & comms: email/SMS templates, PDFs, signatures, watermarking
- integrations: finance (xero/sage), data (vrm/haynespro), maps
- assist: prompts, nudges, guardrails, audit hooks
Narrative reference: docs/onboarding/SMART_ONBOARDING_FLOW.md. Samples: docs/onboarding/sample.ftm.onboarding.json and docs/onboarding/examples/SMART_ONBOARDING.sample.json.
apply & verify
- validate payload → persist under
docs/onboarding/(plus backups/onboarding/) - load:
scripts/onboarding/load_onboarding.py <payload.json>into target env - smoke: quote → approve to job → invoice; PDFs branded; inventory reserve/release; Assist/map basics
- log: /api/internal/logs or OS Observability panels; keep run log + sign-off note
sign-off keeps payload + run results together for reproducibility.
coming soon — assistant UI
we'll wire this page to an OpenAI assistant that walks prospects through the Q&A, streams the draft payload, and lets us export/validate before provisioning. until then, send a note and we'll run it manually.
target: ittd.fleet.logic.ittd.works demo tenant as the first live run.