We emit events like context.assemble, recall.query, mem.upsert, extract.run.
  • Every response includes request_id.
  • Use the dashboard to inspect per-call latency and cost.
Append custom events: 
curl -s https://api.flumes.ai/v0/observability/events \
  -H "Authorization: Bearer $FLUMES_API_KEY" \
  -H "Content-Type": "application/json" \
  -d '{"events":[{"event_type":"context.assemble","event_time":"2024-07-01T12:00:00Z","attrs":{"note":"demo"}}]}'