israeli-employment-contract-reviewer
Pass
Audited by Gen Agent Trust Hub on Jun 28, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted external data (employment contracts) provided by the user, representing a surface for indirect prompt injection attacks.\n
- Ingestion points: Full contract text is collected from the user as the primary object of analysis in
SKILL.mdandSKILL_HE.md.\n - Boundary markers: The instructions do not define delimiters or provide warnings to the agent to isolate the user-provided text from its own instructions.\n
- Capability inventory: The skill configuration in
SKILL.mdspecifies an emptyallowed-toolslist, meaning the agent lacks access to network, file-write, or shell execution tools that could be abused via injection.\n - Sanitization: No input validation, escaping, or sanitization logic is specified for the contract text before it enters the agent's context.
Audit Metadata