review-post-performance

Pass

Audited by Gen Agent Trust Hub on Oct 24, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external social media posts through tool calls like get_post_results. While this creates a surface for indirect prompt injection if post content contains instructions, the risk is minimal as the skill lacks dangerous capabilities like file writing or arbitrary network access.
  • Ingestion points: list_posts_advanced and get_post_results in SKILL.md.
  • Boundary markers: Not present.
  • Capability inventory: No subprocess, file system, or arbitrary network tools are utilized in the provided instructions.
  • Sanitization: Not explicitly implemented in the prompt instructions.
  • [SAFE]: The instructions focus on summarizing analytics data and provide guidance on handling missing data and avoiding invented benchmarks, which aligns with safe and accurate reporting practices.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 24, 2026, 06:54 AM