scenario-analyzer
Pass
Audited by Gen Agent Trust Hub on Mar 27, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by interpolating unvalidated user headlines and external web search content into sub-agent prompts.\n
- Ingestion points: User-supplied news headlines in
SKILL.mdand related news articles collected via theWebSearchtool by thescenario-analystagent.\n - Boundary markers: The prompts utilize Markdown headers to organize data but lack specific delimiters or directives instructing the agents to ignore instructions embedded within the external content.\n
- Capability inventory: The skill possesses the ability to perform web searches and write report files to the
reports/directory via the agent orchestration workflow.\n - Sanitization: No input validation, sanitization, or escaping is performed on the headlines or the external web data before they are processed by the agents.
Audit Metadata