70High
The skill risks RAG poisoning via an external `SK
Behavior Analysis
Mismatch detected — The skill's initial description claims it can retrieve AI products (summaries, to-dos, chapters) from minutes, but later sections explicitly state this functionality is not part of this skill and should be routed to `lark-vc +notes`.
Claims to do
minutes (v1): **CRITICAL — 开始前 MUST 先用 Read 工具读取 [`../lark-shared/SKILL.md`](../lark-shared/SKILL.md)，其中包含认证、权限处理**
Actually does
This skill uses the `lark-cli` tool to interact with Feishu (Lark) Minutes. It can search for minutes using `lark-cli minutes +search`, retrieve basic metadata like title and duration with `lark-cli minutes minutes get`, and download audio/video files or get download URLs using `lark-cli minutes +download`. It also instructs on parsing minute tokens from URLs.
Skill Info
Namelarksuite/cli/lark-minutes
Registrygithub
Version5fa68ccaa04e
PURLpkg:github/larksuite/cli@5fa68ccaa04e?skill=lark-minutes
Stars7,923
Installs57,000
Source
SHA-2560d1ee5164d90...
SHA-1ff205c4c8d35...
MD57586493d920b...
TLSH37e195370d16...
ssdeep96:ZPX/c1YIq...
Size6,845 bytes
Install
This skill has been flagged as potentially malicious. Review the findings below before installing.
Skills.shDocs
npx skills add https://github.com/larksuite/cli --skill lark-minutesAssessments (3)
AI Agent Traps ↗RAG Poisoning via External Configurationhighrag poisoning
The skill explicitly instructs the agent to read an external `SKILL.md` file for critical authentication and permission handling information. If this external file is compromised, it could lead to RAG poisoning, manipulating the agent's security policies or behavior.
90% confidenceline 13
CRITICAL — 开始前 MUST 先用 Read 工具读取 [`../lark-shared/SKILL.md`](../lark-shared/SKILL.md)，其中包含认证、权限处理
Potential Resource Abuse via Unbounded Searchmediumresource abuse
The `minutes +search` command can return up to 200 items per page with no fixed upper limit on total results. This could lead to excessive resource consumption (network, storage, processing) if the agent is instructed to retrieve all results without proper pagination limits.
80% confidenceline 76
`minutes +search` 单次最多返回 `200` 条；结果总数没有固定上限
Description Mismatchmediumdescription mismatch
The skill's initial description claims it can retrieve AI products (summaries, to-dos, chapters) from minutes, but later sections explicitly state this functionality is not part of this skill and should be routed to `lark-vc +notes`.
85% confidence
Description: '4.获取妙记相关 AI 产物（总结、待办、章节）'. Core Scenarios 4: '当用户说...不属于本 skill。应使用 [vc +notes --minute-tokens]...'
Badge
Markdown
[](https://mondoo.com/ai-agent-security/skills/github/larksuite/cli/lark-minutes)HTML
<a href="https://mondoo.com/ai-agent-security/skills/github/larksuite/cli/lark-minutes"><img src="https://mondoo.com/ai-agent-security/api/badge/github/larksuite/cli/lark-minutes.svg" alt="Mondoo Skill Check" /></a>Image URL
https://mondoo.com/ai-agent-security/api/badge/github/larksuite/cli/lark-minutes.svgSecure your AI agents
Skills can read files, run commands, and access credentials. Mondoo helps organizations manage the security risks of AI agent skills across their entire fleet.
- Continuous scanning of skills across all registries
- Policy enforcement before skills reach your agents
- Integration with your existing security workflow