invoice-investigator

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill uses official vendor-provided tools (Zoho Billing) to perform legitimate business tasks. The instructions strictly limit the agent to a 'propose-only' mode, preventing unauthorized modifications to financial records.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from external tool outputs (Zoho Billing API responses), which creates a potential surface for indirect prompt injection if record fields contain malicious instructions.
  • Ingestion points: Data enters the context through ZohoBilling_Get_an_Invoice, ZohoBilling_Search_Customers, ZohoBilling_Get_a_Customer, ZohoBilling_List_all_Invoices, and ZohoBilling_List_all_Payments.
  • Boundary markers: None identified; tool outputs are processed directly without explicit delimiters or warnings to ignore embedded content.
  • Capability inventory: The skill is limited to information retrieval and display; it lacks capabilities for command execution, file system modification, or network exfiltration.
  • Sanitization: No sanitization or validation of the tool output content is performed before rendering the final timeline.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 05:18 AM
Security Audit — agent-trust-hub — invoice-investigator