ask

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a vulnerability where user-provided input is directly interpolated into the execution logic.
  • Evidence: In SKILL.md, the ${ARGUMENTS} variable is placed directly into the prompt structure under 'User Question', which can be exploited to bypass instructions or inject malicious commands into the parallel agents.
  • Ingestion points: Untrusted data enters the agent context through the ${ARGUMENTS} variable and via the findings aggregated from the 'Explore' sub-agents.
  • Boundary markers: The skill does not employ boundary markers, delimiters, or explicit instructions for the agent to ignore potentially malicious content within the search results or user questions.
  • Capability inventory: The skill utilizes the Task tool to orchestrate multiple sub-agents, creating an attack surface where a single injection could influence multiple parallel processes.
  • Sanitization: There is no evidence of sanitization, escaping, or validation of the input arguments or the search findings before they are processed by the synthesis step.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 08:19 AM