← 返回 Skills 市场
69
总下载
1
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install tui-use
功能描述
Operate TUI (terminal user interface) programs through the tui-agent CLI tool. Use this skill whenever the user asks to interact with, automate, control, or...
安全使用建议
Before installing or using this skill: 1) verify you actually have a trusted 'tui-agent' binary available (the registry metadata should have declared this dependency but it doesn't). 2) Do not blindly run the SKILL.md's install advice; the repo URL points to an arbitrary GitHub user and the suggested 'pip install -e tui-agent/' is ambiguous — prefer an official release package or inspect the repository contents before installing. 3) Understand that automating terminal programs lets the tool see everything shown in the terminal (potentially sensitive), and the external tui-agent daemon will spawn PTYs and run commands — audit the code if you will install it. 4) If you proceed, restrict usage to manual invocation (do not enable any always-on/autonomous behaviors for untrusted skills), and consider running the tool in a sandboxed environment until you confirm its safety.
功能分析
Type: OpenClaw Skill
Name: tui-use
Version: 1.0.0
The `tui-use` skill enables an AI agent to launch and interact with arbitrary terminal programs via the `tui-agent` CLI, a high-risk capability involving process execution and screen capture. While the instructions in `SKILL.md` describe legitimate use cases like system monitoring and text editing, the tool's design allows for the execution of any command. Additionally, the skill references an external repository (`github.com/zbflcy/tui-use`) for installation, which poses a potential supply chain risk if the source is not verified.
能力评估
Purpose & Capability
The name/description match the instructions: the skill is an instruction set for controlling TUI programs via a CLI called tui-agent. However, the registry metadata declares no required binaries while SKILL.md explicitly requires a 'tui-agent' CLI (and points to a GitHub repo). The missing declared dependency is an inconsistency: a legitimate TUI automation skill should list the CLI as a required binary or provide a proper install spec.
Instruction Scope
The SKILL.md stays on-topic: it documents starting sessions, capturing screens, sending keys, waiting for text, etc., and does not instruct reading arbitrary system files or exporting credentials. That said, the skill presumes the agent will run arbitrary terminal programs (editors, database clients, monitors) — these programs can expose sensitive data on screen. The instructions also point users to obtain code from a third-party GitHub repo and to run pip install -e tui-agent/, which is an action that downloads and executes external code (outside the skill bundle).
Install Mechanism
There is no install spec in the registry (instruction-only), but SKILL.md tells the user to clone or use a GitHub tree URL and run 'pip install -e tui-agent/'. That guidance is ambiguous (pip install -e expects a local path) and directs installing code from an external GitHub repository (zbflcy/tui-use). Installing arbitrary code from a third-party repo is higher risk and should be declared in install metadata or replaced with a clear, verifiable release URL or package. Because the skill encourages installing/executing external code without provenance, this is a concern.
Credentials
The skill declares no required environment variables, no credentials, and no config paths. The SKILL.md likewise does not request secrets. This is proportionate. Note: automating terminal programs inherently gives the operator (or an agent using this skill) visibility into anything printed to the terminal (which could include secrets), so operational caution is warranted even though no explicit secret access is requested.
Persistence & Privilege
Flags show no 'always: true' and the skill is user-invocable only. Autonomous model invocation is allowed by default but not an additional privilege requested here. The skill does refer to a daemon started by the tui-agent runtime, but that daemon is part of the external tool rather than the skill package itself. No evidence the skill tries to modify other skills or system-wide agent settings.
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 tui-use - 安装完成后，直接呼叫该 Skill 的名称或使用
/tui-use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
v1.0.0
tui-use 1.0.0 – initial release
- Introduces tui-agent CLI tool for automating and controlling terminal user interface (TUI) programs.
- Supports starting, stopping, and managing multiple TUI application sessions.
- Provides commands for screen capture, text typing, special key presses, and clipboard-style paste.
- Enables mouse operations including click and scroll in TUI sessions.
- Adds robust waiting mechanisms for program states and screen stability.
- Documents best practices and example workflows for tools like vim and nano.
元数据
常见问题
TUI USE 是什么？
Operate TUI (terminal user interface) programs through the tui-agent CLI tool. Use this skill whenever the user asks to interact with, automate, control, or... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 69 次。
如何安装 TUI USE？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install tui-use」即可一键安装，无需额外配置。
TUI USE 是免费的吗？
是的，TUI USE 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
TUI USE 支持哪些平台？
TUI USE 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 TUI USE？
由 zbflcy（@zbflcy）开发并维护，当前版本 v1.0.0。
推荐 Skills