lark-crm
Fail
Audited by Snyk on Aug 6, 2026
Risk Level: HIGH
Full Analysis
HIGH W007: Insecure credential handling detected in skill instructions.
- Insecure credential handling detected (high risk: 1.00). The skill tells the agent to extract a base_token from a user-provided URL, cache it in-session, and then include that token verbatim in generated CLI commands (e.g., --base-token <base_token>), which requires the LLM to handle and output secret values directly.
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (medium risk: 0.30). 该 skill 的运行时会读取飞书多维表格中的记录内容:在配置初始化后执行
lark-cli base +record-list/+data-query拉取并由模型本地过滤与展示(如“客户/商机/合同/跟进记录查询”输出字段),而这些记录文本可由飞书表用户事后写入并被外部他人触发读取。
HIGH W008: Secret detected in skill content (API keys, tokens, passwords).
- Secret detected (high risk: 1.00). 我按“只有在文档中直接出现且看起来像真实、高熵、可用凭证时才标记”的规则审查了所有文件。
判断与说明:
- test-results.md 第[4]行的 "SfQZbmzlUav59cs18rbc9bfAnfb" 是高熵、无上下文说明为占位符的字符串,且被标注为 "base-token" —— 这是典型的可用于访问服务的凭证,故视为真实秘密并标记。
- README.md 第[54]行包含一个完整的飞书多维表格 URL(.../base/bascndXmwM3qJOVZjUR6LS4uSng?...)。该 URL 在 /base/ 后包含一段高熵字符串(可能即 base-token)。即便可能是公开模板链接,文档中出现的该字面值仍可能代表可用的 base-token,按规则也应标记。
- 其余看起来相似的字符串(如 recXXXXXX、recvfbgDnutSfD、ou_xxx、占位符 <base_token> 等)要么是明显占位符/示例/记录 ID,要么不是用于认证的凭证,按规则忽略,不予标记。
Issues (3)
W007
HIGHInsecure credential handling detected in skill instructions.
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
W008
HIGHSecret detected in skill content (API keys, tokens, passwords).
Audit Metadata