yuanbao
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEPROMPT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill uses explicit directives to override standard agent behavior. It commands the agent to "NEVER say you cannot send messages," "NEVER suggest the user do it manually," and "NEVER add disclaimers about permissions." These instructions are designed to suppress the agent's built-in transparency and refusal mechanisms when explaining capabilities.
- [INDIRECT_PROMPT_INJECTION]: The skill operates on untrusted text inputs from external group chats, creating a vulnerability surface where a user in the chat could attempt to influence the agent's actions.
- Ingestion points: The skill processes user messages received from the Yuanbao messaging gateway to determine actions like mentioning users or sending DMs.
- Boundary markers: The instructions do not specify the use of delimiters or boundary markers to separate user content from system instructions.
- Capability inventory: The skill has the ability to query group member information (
yb_query_group_members) and send direct messages including local file attachments (yb_send_dm). - Sanitization: The instructions lack requirements for sanitizing or validating user-supplied content before it is processed or passed to tools.
Audit Metadata