qqbot-cron
Pass
Audited by Gen Agent Trust Hub on Mar 15, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill uses a prompt template within the
payload.messagefield that directly interpolates user-provided reminder text ({提醒内容}). - Ingestion points: User-supplied content for reminder descriptions as defined in the interaction logic of
SKILL.md. - Boundary markers: The template does not utilize delimiters or specific instructions to the model to ignore potential commands within the user-supplied string.
- Capability inventory: The skill leverages a
crontool to schedule futureagentTurnactions which execute the generated prompt. - Sanitization: There is no evidence of filtering or escaping of the user content before it is used in the prompt construction.
Audit Metadata