aikey-drama
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data to generate scripts, which presents a surface for potential instruction injection. \n
- Ingestion points: User-provided premises, topics, and initial content used during script development (SKILL.md, Phase 0 and 1). \n
- Boundary markers: Absent; there are no specific delimiters or warnings to ignore instructions that might be embedded in the user's input. \n
- Capability inventory: The skill can write files to the local environment and call external tools for video generation. \n
- Sanitization: The skill does not explicitly mention any sanitization or escaping of user input before processing. \n- [COMMAND_EXECUTION]: The workflow involves the use of scripts for automated content auditing. \n
- Evidence: SKILL.md specifies that banned word checks are performed by running a script ('禁词一律跑脚本扫'), which involves executing code to scan text for prohibited content.
Audit Metadata