green-invoice

Pass

Audited by Gen Agent Trust Hub on Apr 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions and the bundled helper script scripts/green-invoice-client.py facilitate interaction with the service via shell commands and Python's network libraries.
  • [EXTERNAL_DOWNLOADS]: Documentation in references/api-reference.md identifies standard client libraries for Python and PHP, such as the green-invoice package, to assist with API integration.
  • [DATA_EXFILTRATION]: Network communication is directed to authorized service domains api.greeninvoice.co.il and sandbox.d.greeninvoice.co.il. This includes the necessary transmission of API credentials and financial records to the platform.
  • [PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes data from external API responses. 1. Ingestion points: Data is retrieved from the Green Invoice API through scripts/green-invoice-client.py and curl commands. 2. Boundary markers: The instructions lack specific delimiters or warnings to isolate instructions potentially embedded in the API data. 3. Capability inventory: The skill uses network operations, shell command execution, and script execution. 4. Sanitization: The skill does not currently implement sanitization or filtering of the content returned from the API.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 14, 2026, 11:09 AM