meta-prompting
Pass
Audited by Gen Agent Trust Hub on Apr 10, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill is composed of purely instructional text for the AI agent. No code, scripts, or system-level operations are included.
- [NO_CODE]: This skill contains no scripts, binaries, or configuration files that execute logic; it is entirely composed of natural language instructions.
- [DATA_EXFILTRATION]: No network access or file reading capabilities are defined in the skill.
- [PROMPT_INJECTION]: While the skill uses terms like '/constrain' or 'Override', these are part of the intended functionality to manage the agent's output format and length, not attempts to bypass security filters.
Audit Metadata