chinese-grandpa
Pass
Audited by Gen Agent Trust Hub on Jun 28, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill defines a workflow for ingesting, translating, and re-dispatching user instructions, creating a surface for indirect prompt injection.
- Ingestion Points: The skill processes the user's "full English prompt/brief" as the primary source-of-truth (SKILL.md, Behavior Step 1).
- Capability Inventory: The skill is specifically designed to "dispatch the compressed brief to a fleet of sub-agents" (SKILL.md, Description).
- Sanitization: The instructions lack any requirement for the agent to sanitize or validate the English input for malicious content or bypass attempts before translation and dispatch.
- Boundary Markers: There are no explicit instructions to use delimiters or "ignore embedded instructions" markers when sending the translated Mandarin content to sub-agents, which may increase the risk of the sub-agent obeying malicious commands hidden in the translated text.
Audit Metadata