customer-360

Pass

Audited by Gen Agent Trust Hub on Aug 6, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted conversation history and customer metadata retrieved from Intercom, which creates a potential surface for indirect prompt injection.\n
  • Ingestion points: Data enters the agent's context through the output of the search_conversations and get_conversation tools mentioned in the Workflow section of SKILL.md.\n
  • Boundary markers: The instructions do not define specific delimiters or isolation protocols to distinguish between agent instructions and retrieved customer data.\n
  • Capability inventory: The skill uses data retrieval tools (search_contacts, get_contact, search_conversations, get_conversation) but does not include capabilities for file system modification, shell command execution, or external network communication to non-vendor domains.\n
  • Sanitization: No explicit sanitization or filtering logic is provided to validate the contents of the retrieved Intercom data before processing.\n
  • Assessment: This risk factor is inherent to the skill's primary function of summarizing customer interactions and does not indicate malicious intent.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 6, 2026, 09:30 PM