invoice-organizer

Pass

Audited by Gen Agent Trust Hub on Mar 11, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection by processing external data from untrusted invoices.
  • Ingestion points: Scans and reads PDF, JPG, and PNG files from the local file system (SKILL.md).
  • Boundary markers: No explicit delimiters or safety instructions are provided to separate document content from the agent's logic.
  • Capability inventory: The agent can execute shell commands (find, mkdir, cp, mv) and generate CSV files.
  • Sanitization: No sanitization or validation of extracted text is mentioned before it is used in filenames or summary reports.
  • [COMMAND_EXECUTION]: The skill relies on shell commands to perform file discovery and organization.
  • Evidence: Instructions specify the use of find, mkdir, cp, and mv commands for scanning, creating directories, and moving/copying files.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 11, 2026, 04:45 PM