israeli-employment-contracts
Pass
Audited by Gen Agent Trust Hub on Aug 21, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides numerous reference links to official Israeli government domains (gov.il, btl.gov.il) and authoritative legal information portals (kolzchut.org.il, nevo.co.il). These are recognized as well-known and trusted services for legal documentation and do not contribute to a malicious verdict.
- [COMMAND_EXECUTION]: The skill instructions direct the agent to execute a bundled Python script (
generate_contract.py) to automate the generation of contracts. This script is local to the skill, performs standard arithmetic for benefit calculations, and does not exhibit any suspicious side effects such as unauthorized network access or system modification. - [INDIRECT_PROMPT_INJECTION]: The skill exposes a surface for indirect prompt injection by requesting user input (such as names and job titles) and interpolating it into a generated contract. * Ingestion points: Employee details requested in SKILL.md for script parameters. * Boundary markers: Absent. * Capability inventory: The
generate_contract.pyscript includes the capability to write the generated contract to a local file. * Sanitization: Absent; however, as the primary purpose of the skill is document generation based on user input, this is considered an inherent functional risk rather than a malicious vector.
Audit Metadata