al
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill includes instructions that explicitly attempt to override the agent's internal operational logic and reasoning capabilities.
- Evidence: 'Mandatory: Do not use your plan mode. Execute the AAD protocol from alignfirst.'
- This pattern attempts to bypass the agent's standard planning constraints in favor of a specific external protocol.
Audit Metadata