Security audit
Security checks for vulnerabilities and agentic risk
This skill openly performs user-directed read and edit operations on existing DingTalk documents and the inspected code matches that limited purpose.
Install only if you intend to let this skill use your DingTalk enterprise app credentials to read and change existing documents. Treat update and delete-block requests as potentially destructive, confirm the document and block target first, and avoid configuring the local debug operator ID in production.
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
在 OpenClaw 环境中配置：
# ~/.openclaw/.env
DINGTALK_CLIENTID=dingxxxxxx
DINGTALK_CLIENTSECRET=your_secret
Referenced artifact was not completely inspected
使用同目录的 `doc-enterprise.js` 操作用户通过 URL 指定的已有钉钉文档。
Referenced artifact was not completely inspected
使用同目录的 `doc-enterprise.js` 操作用户通过 URL 指定的已有钉钉文档。
Referenced artifact was not completely inspected
将相对路径按 `SKILL.md` 所在目录解析，并使用脚本绝对路径：
This markdown file documents commands that can overwrite an entire document body and delete content blocks, which are safety-critical operations affecting user data. Although the capabilities are listed, there is no explicit warning or caution telling users these actions are destructive or should be confirmed before use.
The skill requires access to environment secrets and networked API operations, but it does not declare an explicit tool scope such as permissions or allowed-tools. That creates an authorization-boundary weakness: the runtime may permit broader capabilities than reviewers or policy systems can easily verify, making secret use and outbound calls less transparent and harder to constrain. In this context, the skill edits enterprise DingTalk documents on behalf of a user, so undeclared capability scope increases the risk of unintended data access or modification if the skill is invoked incorrectly or extended unsafely.