saudi-legal
Pass
Audited by Gen Agent Trust Hub on Jul 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions to execute
curlcommands that interact with the Najiz and Wathq API endpoints. These commands incorporate user-controlled variables such as{caseNumber},{requestNumber},{deedNumber}, and{number}directly into the URL path or query parameters. - [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection due to the way it processes untrusted user data.
- Ingestion points: Untrusted user input is accepted for various legal identifiers (case, deed, and agency numbers) used in API requests defined in
SKILL.md. - Boundary markers: Absent; there are no instructions or delimiters provided to ensure the agent treats these inputs strictly as data and ignores any potentially embedded instructions.
- Capability inventory: The skill utilizes shell command execution (
curl) and network access to communicate with external API services. - Sanitization: Absent; the skill does not include instructions for validating, escaping, or filtering user-provided inputs before they are used to construct and execute shell commands.
Audit Metadata