← 返回 Skills 市场
1475
总下载
0
收藏
5
当前安装
2
版本数
在 OpenClaw 中安装
/install taskboardai-skill
功能描述
Manage tasks and projects using the TaskBoardAI Kanban system. Includes MCP server integration.
安全使用建议
Before installing: (1) Confirm you trust the TaskBoardAI package source — the MCP server the skill runs is provided by that npm package and will execute code on your system. (2) Ensure you have node installed and decide where the kanbanMcpServer.js file will live; the SKILL.md hardcodes a macOS/Homebrew path which may not match your system. (3) Expect the skill to autonomously create and immediately start tasks and to use other tools (web_search, etc.); if you do not want this behavior, do not enable autonomous invocation or modify AGENTS.md triggers. (4) Ask the skill author to declare required binaries and any environment variables (API keys) formally in the skill metadata so you can audit what credentials are needed. (5) If uncertain, test in an isolated/sandbox environment and inspect the installed TaskBoardAI package contents before running the MCP server.
功能分析
Type: OpenClaw Skill
Name: taskboardai-skill
Version: 1.0.1
The taskboardai-skill bundle is a legitimate integration for managing Kanban boards via the Model Context Protocol (MCP). The SKILL.md and README.md files define a clear workflow for task lifecycle management (Create, Execute, Block, Complete) and point to a local Node.js execution path for the TaskBoardAI server. No evidence of data exfiltration, malicious execution, or harmful prompt injection was found.
能力评估
Purpose & Capability
The SKILL.md and README clearly depend on the external TaskBoardAI npm package and on running node against a specific script path (/opt/homebrew/lib/node_modules/taskboardai/server/mcp/kanbanMcpServer.js). The registry metadata, however, declares no required binaries, no install steps, and no environment variables — that's inconsistent. Requiring the TaskBoardAI package and node is reasonable for a TaskBoard integration, but those requirements should be declared by the skill.
Instruction Scope
Runtime instructions tell the agent to spawn a node MCP server process and to immediately move newly created tasks into 'In Progress' and to 'perform the work' using available tools (web_search, etc.). That gives the agent broad, autonomous authority to run processes and use other tool integrations. The SKILL.md also references missing API keys as a blocker but does not declare what credentials or config the MCP server needs. This scope is broader than the metadata indicates and may cause unexpected autonomous actions.
Install Mechanism
There is no install spec in the registry (the skill is instruction-only), but the README instructs users to run `npm install -g taskboardai` and to clone the repo. Relying on a global npm package is a moderate risk: the code executed when the MCP server is launched comes from that package. The skill does not pin or verify the package source in metadata, and the SKILL.md hardcodes a Homebrew-style path which may not exist on non-macOS systems.
Credentials
The skill declares no required environment variables, yet the instructions mention 'missing API key' as a blocker and rely on an external package that likely needs configuration/credentials. Absence of declared env vars or primary credential is a mismatch — you may need to supply secrets later, and the skill gives no guidance about what and where to set them.
Persistence & Privilege
The skill is not always-enabled and does not request special system-wide privileges. However, it instructs the agent to autonomously start tasks and immediately execute them using other tools; combined with the ability to spawn a node process, this raises operational risk if you do not want autonomous task execution. (Autonomous invocation itself is normal for skills and is not flagged alone.)
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install taskboardai-skill - 安装完成后，直接呼叫该 Skill 的名称或使用
/taskboardai-skill触发 - 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
v1.0.1
- bugfix
v1.0.0
Initial release, TaskboardAI github: https://github.com/TuckerTucker/TaskBoardAI.git
元数据
常见问题
TaskboardAI Skill 是什么？
Manage tasks and projects using the TaskBoardAI Kanban system. Includes MCP server integration. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 1475 次。
如何安装 TaskboardAI Skill？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install taskboardai-skill」即可一键安装，无需额外配置。
TaskboardAI Skill 是免费的吗？
是的，TaskboardAI Skill 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
TaskboardAI Skill 支持哪些平台？
TaskboardAI 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 TaskboardAI Skill？
由 hyddd（@hyddd）开发并维护，当前版本 v1.0.1。
推荐 Skills