Signature scanners are fast but blunt: they flag patterns, which sometimes catches legitimate code (a false positive) and sometimes misses cleverly-disguised malware. PowerSEC adds an AI second opinion on top of the deterministic scanner.
Second-opinion malware triage
When the scanner flags a file, PowerSEC can ask an AI model to review the specific evidence — the matched signals, decoded strings, and how the file differs from the known-good version of that plugin (our PluginBaseline hash system) — and return a plain-English verdict: likely clean, needs review, or likely malicious, with the reasons and a recommended next step. This helps you confidently dismiss false positives and prioritise real threats. You'll find it on a site's detail page under the incident Triage view.
Incident summaries
For an active incident, PowerSEC can generate a clear summary — what happened, the safest order to respond, and the recommended actions — so you're not piecing together raw alerts under pressure.
Privacy & safety
- Sensitive data (credentials, API keys, emails) is redacted before anything is sent for analysis.
- The AI is a second opinion — the deterministic scanner and your own judgement remain the source of truth. PowerSEC never auto-deletes files based on an AI verdict.
- Per-site daily limits keep usage predictable.
Note: AI features are activated per account. If your verdicts show "AI unavailable", AI assistance is currently turned off for your account and only the deterministic checks are running.