lisa-repair-intake
Pass
Audited by Gen Agent Trust Hub on Aug 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill explicitly instructs the agent to override standard safety protocols by suppressing confirmation prompts ('do not ask whether to proceed once a queue is known'). This reduction in user oversight increases the risk of automated execution of unintended or malicious commands.- [PROMPT_INJECTION]: The skill acts as an ingestion point for untrusted caller arguments which are passed directly to a secondary execution contract ('plugins/lisa/skills/repair-intake/SKILL.md'). There are no defined boundary markers or sanitization steps to prevent these arguments from influencing the agent's behavior in the downstream skill cycle.
Audit Metadata