← 返回 Skills 市场
longsasasasasa

meego-skill

作者 LONGSASASASASA · GitHub ↗ · v2.2.0 · MIT-0
cross-platform ⚠ suspicious
99
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install meego-skill
功能描述
飞书项目(Meego)全能力技能。提供工作项查询、状态流转、团队管理、工时统计、评论协作等全部能力。当用户提到以下场景时激活:查询在途工作项、查缺陷/任务/需求详情、查团队成员、查工时记录、添加评论、创建工作项、修改工作项字段、查询工作项流转状态、查成员排期、查询视图和图表、查询工作项字段配置、查询项目空间信息、...
安全使用建议
Do NOT paste your ~/.mcporter/credentials.json (or any OAuth tokens) into chat. The SKILL.md explicitly asks you to paste credentials for the assistant to write into /root/.mcporter/credentials.json — that creates a direct credential-exfiltration risk. Safer options: (1) perform the npx auth step directly on the target server or on a secure admin workstation so tokens are written locally and never shared in chat; (2) if you must use this skill, create a scoped/limited service account or ephemeral token with minimal permissions and rotate it immediately; (3) verify the npm package @lark-project/meego-mcporter provenance and audit its code before installing; (4) prefer secure copy (scp/sftp) or a secrets manager to move credentials to servers rather than pasting into chat; (5) if you already shared tokens here, rotate/revoke them immediately. The skill's functionality itself matches its description, but the credential-handling instructions are unsafe — treat them as a blocking concern unless you change to a secure transfer workflow.
功能分析
Type: OpenClaw Skill Name: meego-skill Version: 2.2.0 The meego-skill bundle is a comprehensive integration for Feishu Meego project management, utilizing the `meego-mcporter` CLI tool via the Model Context Protocol (MCP). The documentation in SKILL.md provides transparent instructions for OAuth authorization and tool configuration, including a step where the user provides credentials for the AI to save to `/root/.mcporter/credentials.json`. While the skill involves sensitive token handling and shell command execution, all behaviors are strictly aligned with its stated purpose of managing work items, schedules, and team data within the Feishu ecosystem, with no evidence of malicious intent or unauthorized data exfiltration.
能力评估
Purpose & Capability
Name/description match the provided instructions and references (Meego / 飞书 项目 via an MCP tool). However the registry metadata declares no required binaries or env vars while the SKILL.md explicitly requires Node/npm and the meego-mcporter/mcporter tools—an inconsistency between metadata and runtime requirements (metadata omission, not necessarily malicious).
Instruction Scope
The SKILL.md instructs the user to run npm installs and OAuth locally (expected), but crucially tells the user to cat their local OAuth credentials and paste the full JSON into the chat so the AI will 'automatically write' it into the server's /root/.mcporter/credentials.json. That directs the agent to receive and store sensitive tokens via chat (credential transfer/exfiltration) and to perform file writes on a server path; this goes beyond normal scoped API usage and materially raises data-exposure risk.
Install Mechanism
No formal install spec exists (instruction-only), and the instructions rely on npm installs and npx usage (@lark-project/meego-mcporter and mcporter). Using npm/npx is expected for this tooling, but global installs and npx fetches should be reviewed (package provenance, audit). There is no evidence of arbitrary URL downloads or unusual extract/install steps.
Credentials
The skill requests no declared env vars in metadata but the flow requires OAuth tokens saved in ~/.mcporter/credentials.json. The instructions explicitly ask users to paste the full credentials JSON into chat for the assistant to transfer to the server. That asks for highly sensitive secrets via an insecure channel (chat), which is disproportionate and unnecessary — alternatives exist (run auth directly on the server, use secure file transfer, or provide limited/ephemeral tokens).
Persistence & Privilege
While the skill is not marked always:true and allows normal autonomous invocation, the instructions direct writing credentials to /root/.mcporter/credentials.json and global npm installs on server—operations that grant long-lived access to the user's Meego account if performed. Having the assistant handle raw tokens in-chat increases persistent access risk.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install meego-skill
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /meego-skill 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v2.2.0
meego-skill 2.2.0 - 提供飞书项目(Meego)全流程支持,包括工作项查询、流转、团队管理、工时统计、评论、创建与修改等全场景能力。 - 面向新手用户,详细分步指引每项操作;面向系统对接,所有接口均含参数、示例和异常处理说明。 - 全面重构和完善文档,涵盖依赖环境、安装步骤、OAuth 授权流程和服务器配置的完整自助指引。 - 新增各类场景的标准命令行调用示例,支持 URL 自动解析、分页和字段查询等高级用法。 - 明确工具链依赖与校验方式,便于环境快速配置和错误自查。
元数据
Slug meego-skill
版本 2.2.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

meego-skill 是什么?

飞书项目(Meego)全能力技能。提供工作项查询、状态流转、团队管理、工时统计、评论协作等全部能力。当用户提到以下场景时激活:查询在途工作项、查缺陷/任务/需求详情、查团队成员、查工时记录、添加评论、创建工作项、修改工作项字段、查询工作项流转状态、查成员排期、查询视图和图表、查询工作项字段配置、查询项目空间信息、... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 99 次。

如何安装 meego-skill?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install meego-skill」即可一键安装,无需额外配置。

meego-skill 是免费的吗?

是的,meego-skill 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

meego-skill 支持哪些平台?

meego-skill 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 meego-skill?

由 LONGSASASASASA(@longsasasasasa)开发并维护,当前版本 v2.2.0。

💬 留言讨论