Production

Turn it on#

WizeLens is off in production until you turn it on. Set these on each service and restart it:

WIZELENS_ENABLED=true
WIZELENS_HUB=https://app.wizelens.com
WIZELENS_HUB_TOKEN=wl_ik_...                   # one ingest key per service
WIZELENS_SAMPLE_RATE=0.25                      # optional, see Sampling

No code change is needed. Enabled = true (.NET) or enabled: true (Node) does the same as WIZELENS_ENABLED.

Send-only#

Outside development the add-on is send-only. It captures every request, query, log, and browser span, and sends them to the cloud or hub. It serves no dashboard, read API, or MCP, so nobody can read traces from the app itself.

Path under /_wizelens	Send-only
POST api/client-spans (browser spans)	Answers
Dashboard, api/info, api/traces, metrics, MCP	404

Traces stay in memory only, until they are sent. The cloud gets no address to read from the app. "Outside development" means outside the Development environment (.NET) or with NODE_ENV=production (Node, Next.js).

To serve the dashboard anyway, for example on a staging server behind a VPN, set WIZELENS_DASHBOARD=true (Dashboard = true / dashboard: true).

The browser#

Hide the HUD, and keep sending page loads, React renders, and browser errors:

<WizeLensProvider hud={false}>
  <App />
</WizeLensProvider>

With hud={false} the browser never reads traces back, so it works with a send-only server.

One key per service#

Each server sends a batch every second. The cloud takes at most 1200 calls per key per minute, so give each service its own ingest key, all in the same project. One project keeps a trace whole when it crosses services.