Aurex watches LLM calls locally, hashes prompts for zero-PII telemetry, and optionally syncs metadata to your Aurex project. Budget guards and provider resilience run in-process without adding latency to every call.
Start at /llms.txt for task → URL routing. Fetch one page per concern (e.g. /docs/reference/config, /docs/reliability/policy) — avoid loading hub pages for implementation detail.
Install the Python or Node SDK, configure API keys, and log your first event.
Serverless flush, filesystem fallbacks, cloud sync resilience, and ingest rate limits.
Choose patch vs wrap vs manual logging based on your observability stack.
Focused pages for config, guards, providers, health, and CI — one topic per URL.
Provider-call retry, fallbacks, observe/enforce modes, and loss_avoided accounting.