Request Logs
The Request Logs component of your dashboard provides unparalleled visibility into your AI traffic passing through the EdgeMask proxy. Here you can monitor latencies, caching hits, PII redactions, and overall request volume.
What gets logged?
To ensure peak observability without compromising data privacy, EdgeMask logs valuable metadata about each request while rigorously preventing raw user prompts and model completions from being stored on our servers.
Metrics
Timestamps, HTTP status codes, latencies, model details, token counts, and costs are tracked for performance analysis.
Cache Status
Instantly identify whether a response was a HIT or MISS against the Semantic Edge Cache.
Security Enforcement
View detected PII categories (e.g., Email, Credit Card) per request without exposing the underlying confidential data.
Zero Data Retention Policy: EdgeMask guarantees that raw contents of your LLM prompts are completely transient. They exist only in RAM during the routing process and are never written to any persistent database.