Skip to content

hosted.md

Latest commit

 

History

History
43 lines (36 loc) · 1.8 KB

hosted.md

File metadata and controls

43 lines (36 loc) · 1.8 KB
Raw

Hosted mode

VEIL is local-first by default. Hosted mode adds browser sessions and per-session twin directories while keeping local mode unchanged.

VEIL_HOSTED=1 \
VEIL_SESSION_SECRET="$(openssl rand -hex 32)" \
VEIL_HOSTED_ROOT=/srv/veil-hosted \
OPENAI_REQUIRE_USER_KEY=1 \
CHAT_PROVIDER=openai \
HOST=0.0.0.0 \
npm start

How it works:

  • The server issues a signed veil_session HttpOnly cookie.
  • The first request creates $VEIL_HOSTED_ROOT/users/<session>/twins/default/data.
  • Until that session has a built twin, / serves the same setup UI used by npm run init. The visitor draws an AOI, chooses optional national layers, and the server builds that session's twin.
  • /data/*, chat MCP tools, annotations, simulation, and geospatial uploads use the session twin directory.
  • The chat panel still uses bring-your-own OpenAI keys. With OPENAI_REQUIRE_USER_KEY=1, the server never falls back to an operator key.
  • Live telemetry, building placement writes, and survey uploads are disabled in hosted mode until they have product-grade account and device ownership rules.

Do not put a personal or private twin in VEIL_HOSTED_TEMPLATE_DATA_DIR for a public deployment. That variable is optional and should only point at a deliberately shareable starter dataset.

Users can add their own geospatial layers from the Layers sidebar. The upload endpoint saves all dropped files together, then runs scripts/add_layer.py, which reads GeoTIFF/vector data through GDAL/OGR, reprojects to the twin CRS, clips to the terrain footprint/AOI grid, emits drape assets, and updates the Atlas catalog.

For production, put the Node server behind HTTPS, set a stable VEIL_SESSION_SECRET, back up VEIL_HOSTED_ROOT, and enforce reverse-proxy request size/rate limits in addition to VEIL_LAYER_UPLOAD_MAX_BYTES.