The Agent Skills Directory

[DATA_EXFILTRATION]: The skill accesses chat messages and sensitive credentials via environment variables (FEISHU_APP_ID, FEISHU_APP_SECRET). All network operations are strictly limited to open.feishu.cn, the official domain for Feishu/Lark API services. No data is transmitted to unauthorized or unknown external domains.
[COMMAND_EXECUTION]: The script check_missed_mentions.py contains no calls to dangerous functions such as eval(), exec(), or subprocess modules that would allow arbitrary command execution.
[CREDENTIALS_UNSAFE]: The skill follows security best practices by retrieving API secrets from the environment rather than hardcoding them within the source code.
[PROMPT_INJECTION]: The SKILL.md file contains standard instructional language for managing Feishu groups and does not include any patterns intended to bypass agent safety filters or override system instructions.

feishu-chat-monitor