ideal-customer-profile-matching

Pass

Audited by Gen Agent Trust Hub on Mar 18, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill describes an architecture for processing untrusted conversation data to influence agent behavior, which creates a surface for indirect prompt injection.
  • Ingestion points: The update_icp_during_conversation function in SKILL.md processes untrusted prospect messages via the message parameter.
  • Boundary markers: The logic snippets do not include delimiters or instructions for the agent to ignore commands that may be embedded in the prospect's input.
  • Capability inventory: The skill uses extracted data to drive automated messaging in customize_message_for_icp and perform automated lead routing in route_by_icp.
  • Sanitization: There is no evidence of input validation or filtering before the conversation signals are used to update the system state.
  • [EXTERNAL_DOWNLOADS]: The Python code snippets reference the well-known scikit-learn library for machine learning functionality.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 18, 2026, 10:46 PM