full-output-enforcement
Pass
Audited by Gen Agent Trust Hub on Mar 26, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill uses directive language and override markers ('Overrides default LLM truncation behavior', 'No exceptions', 'Hard failures') to modify the agent's core response logic. While intended for utility to ensure output completeness, these instructions utilize patterns characteristic of prompt injection aimed at bypassing standard operational constraints.
Audit Metadata