israeli-insurance-duplication-checker
Pass
Audited by Gen Agent Trust Hub on Aug 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it requires the agent to ingest and process data from untrusted external sources such as insurance reports and bank statements. * Ingestion points: Step A of the audit workflow involves reading Har HaBituach files, pension clearing house reports, payslips, and credit card statements. * Boundary markers: The skill does not provide explicit delimiters or instructions to the agent to disregard potential instructions embedded within the processed document data. * Capability inventory: The agent is tasked with using the kolzchut MCP tool and generating Hebrew-language cancellation templates based on audit findings. * Sanitization: There are no specific instructions to filter or sanitize the content of the ingested files.
- [SAFE]: The skill includes strong security-positive instructions, specifically directing users to official government domains like harb.cma.gov.il and providing warnings about fraudulent lookalike domains.
Audit Metadata