meeting-notes
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill consists entirely of natural language instructions and Markdown templates. There are no executable scripts (Python, JavaScript, shell) or configuration files that invoke local system commands.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted meeting data (Step 1: 获取会议信息). While this represents an attack surface for indirect prompt injection (e.g., hidden instructions in transcripts), the skill lacks any impactful capabilities such as file system writing, tool invocation, or network access to exploit such an injection. No boundary markers or sanitization are currently specified between user data and prompt instructions.
- [REMOTE_CODE_EXECUTION]: No remote code execution patterns or package installations were found.
- [DATA_EXFILTRATION]: The skill clarifies in its boundaries (能力边界) that it cannot automatically send emails or access external project management tools (Jira, Feishu), preventing automated data exfiltration via API or network calls.
Audit Metadata