workload-analysis

Pass

Audited by Gen Agent Trust Hub on Apr 1, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill interacts with the Vitally API to retrieve account metadata, including current MRR, health scores, and contract intervals.
  • [COMMAND_EXECUTION]: The skill executes SQL queries against the internal PostHog billing database to extract SDK event volumes and billable metrics.
  • [PROMPT_INJECTION]: The skill establishes an indirect prompt injection surface by ingesting external data (from Vitally and SQL) and interpolating it into a generated React component; this is typical for data visualization skills and is consistent with the primary workload analysis purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 1, 2026, 05:26 PM