playbook-geo-newsjacking

Pass

Audited by Gen Agent Trust Hub on Apr 18, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill includes a master prompt template that requests external news content to be pasted into the prompt. This creates a surface for indirect prompt injection where instructions hidden within news source material could attempt to hijack the resulting output. Given this is a manual playbook for human-in-the-loop workflows, the impact is confined to content generation.
  • Ingestion points: Section 3 placeholder for news headlines and paragraphs.
  • Boundary markers: Absent (no delimiters or ignore-instructions are specified in the template).
  • Capability inventory: None (the skill is informational and focuses on text generation).
  • Sanitization: Absent (processing of external text relies on the user's discretion).
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 18, 2026, 01:41 AM