prospect

Pass

Audited by Gen Agent Trust Hub on Apr 23, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests and processes untrusted data from the internet to drive its analysis and decision-making processes.
  • Ingestion points: External research data is fetched using WebSearch and WebFetch in Phase 2.1 and then passed as context to sub-agent prompts during the intelligence brief phase.
  • Boundary markers: The instructions do not define clear delimiters (like XML tags or markdown blocks) or include explicit system warnings to the model to ignore instructions contained within the fetched research data.
  • Capability inventory: The skill environment possesses broad capabilities, including the Bash tool for shell operations and file system tools (Write, Edit, Glob, Grep) which could be targeted by a successful injection.
  • Sanitization: There is no evidence of sanitization, filtering, or validation of the content retrieved from external sources before it is interpolated into agent prompts.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 23, 2026, 11:52 PM