arithmetic
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of instructional text for generating math problems based on user-provided parameters (grade level, problem type, and quantity).
- [NO_CODE]: There are no scripts, binaries, or command-line executions associated with this skill. It relies purely on the AI's natural language generation capabilities.
- [DATA_EXPOSURE]: No hardcoded credentials, sensitive file paths, or network operations were identified.
- [OBFUSCATION]: The content is clear and uses standard Chinese and mathematical notation without any hidden characters or encoded strings.
- [INDIRECT_PROMPT_INJECTION]: While the skill processes user input (to determine the types of math problems to generate), it lacks high-risk capabilities like file writing or network access, making the exploitation surface negligible.
Audit Metadata