Security audit
Security checks for vulnerabilities and agentic risk
This skill is a documented Feishu messaging helper that sends user-selected text and files through OpenClaw, with no hidden code, persistence, or unrelated access found.
Install only if you intend to let the agent send Feishu messages and chosen files. Before sending screenshots, documents, or logs, confirm the recipient ID and remove secrets, tokens, personal data, customer data, and internal details that should not leave your environment. Test bulk sends with one recipient first.
Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).
mkdir -p ~/.openclaw/workspace/temp/
rm ~/.openclaw/workspace/temp/*
### 2. 消息模板
The README instructs users to send messages, images, and files to Feishu, but it does not clearly disclose that local content will be transmitted to an external service or remind users to verify consent and data sensitivity before sending. This can lead to accidental exfiltration of confidential files or personal data, especially because the examples normalize sending local media paths with minimal friction.
The documentation consistently uses Chinese for headings, instructions, examples, and troubleshooting, which effectively forces a specific language on users without an opt-in or alternative. The file does not state that the skill is region-specific or provide another language option.
The skill explicitly encourages sending screenshots, documents, and packaged logs to an external messaging platform but does not warn about secrets, credentials, personal data, or internal system details that may be embedded in those files. In an agent setting, this increases the risk of accidental exfiltration of sensitive information because users may follow the documented workflow without performing any content review or recipient validation.
Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.
解决:
# 检查文件权限
chmod 644 ./file.ext
# 或重新复制文件
cp /path/to/file.ext ./
Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.
# 创建临时文件夹
mkdir -p ~/.openclaw/workspace/temp/
# 使用后清理
rm ~/.openclaw/workspace/temp/*
SQP-3 applies to natural-language policy issues in any file type. The operational description and guidance are written in Chinese only, and the file does not indicate that this is a region-specific skill or offer an alternative language, which can amount to a forced language choice without opt-in.
The bulk messaging loop demonstrates mass outbound communication without requiring recipient verification, dry-run confirmation, rate limiting, or environment scoping. This can lead to unintended spam, misdelivery to the wrong users, or operational disruption if copied directly into automation.
The manifest description is written entirely in Chinese ('飞书消息发送 - 支持文本、图片、文件'), which indicates a language-specific presentation without any indication that users can choose another language or that the skill is limited to a Chinese-speaking context. This can conflict with language/locale policy requirements when no opt-in or justification is provided.
No suspicious patterns detected.