financial-document-parser

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [Indirect Prompt Injection] (LOW): The skill processes untrusted financial documents (PDFs, images) which are a known surface for indirect prompt injection attacks where malicious instructions are embedded in the data.
  • Ingestion points: User-provided invoices, receipts, bank statements, and tax documents in PDF or image format.
  • Boundary markers: Absent. The instructions do not define delimiters or warnings for the agent to ignore instructions found within the parsed documents.
  • Capability inventory: No code execution or network capabilities are defined within the skill scripts (none provided).
  • Sanitization: Partial. The skill instructs the agent to mask sensitive account numbers, but does not provide instructions to filter or sanitize executable-like text from document content.
  • [No Code] (SAFE): The skill consists entirely of instructional text and does not provide any scripts, binaries, or configuration files that execute logic on the host system.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:34 PM