diagnosing-missing-recordings

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: Analysis of the skill instructions and reference files reveals no malicious patterns. The tool usage is consistent with the stated purpose of diagnosing session capture issues within the PostHog platform. The diagnostic logic is well-defined and uses appropriate platform-specific signals to provide actionable recommendations.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes external data from PostHog event properties. Ingestion points: Event properties queried via posthog:execute-sql in SKILL.md. Boundary markers: None. Capability inventory: Querying and analyzing recording data. Sanitization: The diagnostic logic maps specific enum-like properties (e.g., $recording_status, $session_recording_start_reason) to predefined verdicts, which prevents arbitrary strings from being interpreted as instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 03:22 PM
Security Audit — agent-trust-hub — diagnosing-missing-recordings