Security audit
Security checks for vulnerabilities and agentic risk
This TAPD skill is a straightforward workplace project-management helper with disclosed TAPD and requested Enterprise WeChat actions, but users should still confirm any update or scheduled notification before it runs.
Install this only where the TAPD MCP and Enterprise WeChat permissions match your intended workspace. Confirm before allowing it to change TAPD states, add comments, send messages, or create recurring notifications, especially where project or personal work data is sensitive.
The activation description is broad enough to apply to many TAPD-related requests without precise boundaries on when the skill should or should not act. Overly broad triggering can cause the agent to invoke TAPD operations or related tools in situations where the user only wanted discussion, analysis, or read-only assistance, leading to unnecessary access or unintended state changes.
All natural-language instructions and examples in the skill are written in Chinese, with no indication that the user can choose another language or that the skill is intentionally limited to a Chinese-only audience. This can violate language or locale policy when a skill implicitly forces one language without user opt-in.
The skill is scoped as TAPD project/query-update functionality, but it also instructs the agent to send Enterprise WeChat messages. This introduces an out-of-scope side-effect capability that could be triggered during a TAPD-related interaction, increasing the risk of data exfiltration, unauthorized notifications, or cross-system action without clear user intent boundaries.
No suspicious patterns detected.
