Audit and access logs
See every query, log search and file read TraceX made in the data access log, every sign-in and setting change in the audit log, and what the AI safeguards stopped.
“Trust, but verify.” TraceX keeps three logs so you can see exactly what happened: what it read, what people changed, and what its safeguards stopped. All are under Security.
Data access log
Section titled “Data access log”Security → Data access lists every query, log search and file read TraceX made of your systems, with:
- the system it read (which database, log store or repository),
- what it ran or read,
- the investigation or question it was for,
- when.
Export it as CSV for your own review or SIEM. Use it to answer questions like
“did TraceX ever read the payments table?”.
Audit log
Section titled “Audit log”Security → Audit log records sign-ins, setting changes, connections and data exports, with who, when and from where. Filter by action. The audit log covers a year on the Team plan.
Typical uses: checking who connected a database, who changed a security setting, or reviewing sign-ins after someone leaves.
AI security events
Section titled “AI security events”Security → AI safeguards lists what TraceX’s AI safeguards stopped: prompt extraction attempts, jailbreaks, hidden instructions in content it read, and secrets. It is part of the Team plan.
A hidden instruction is usually text in a customer email or ticket that tries to make an AI do something else (“ignore your instructions and…”). Seeing these logged is expected and means the safeguards worked.