exploring-live-traffic

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is purely instructional markdown designed to help an AI agent interpret user requests regarding PostHog's real-time web analytics and generate the appropriate query structures (JSON) for product analytics insights.
  • [SAFE]: No evidence of prompt injection or attempts to bypass safety guardrails were found in the instructions.
  • [SAFE]: The skill does not perform any external network requests, file system modifications, or command executions. It operates entirely within the context of the PostHog platform's expected behavior.
  • [SAFE]: Data processing is limited to constructing PostHog-specific TrendsQuery JSON objects based on user-provided parameters like hostnames or bot names. This is standard functionality and lacks dangerous capabilities like subprocess execution or arbitrary network access.
  • [SAFE]: No obfuscation techniques, hidden URLs, or hardcoded credentials were detected.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 06:42 AM
Security Audit — agent-trust-hub — exploring-live-traffic