← 返回 Skills 市场
tanjinlimkelvin-dot

Prompt Buttons

作者 tanjinlimkelvin-dot · GitHub ↗ · v1.0.2 · MIT-0
cross-platform ✓ 安全检测通过
149
总下载
0
收藏
1
当前安装
1
版本数
在 OpenClaw 中安装
/install prompt-buttons
功能描述
A comprehensive helper skill that wraps agent prompts with short, consistent tappable button menus (Yes/No, single digits, or small option sets) so users can...
安全使用建议
This package appears to be a small, focused helper for building and sending button-prompts and is internally consistent overall. Before installing: (1) review and confirm you trust the source/author (no homepage or repo is provided); (2) follow the SKILL.md installation steps manually and avoid blindly running scripts that write to /data or enable the skill in openclaw.json unless you control the agent; (3) ensure callback_data never contains secrets or tokens — map short action keys to server-side state as recommended; (4) note the minor discrepancy: documentation mentions allowing up to 10-character labels in some cases, but the code rejects labels longer than 3 characters — update docs or code as needed for your use; (5) test the skill in a staging agent to confirm messageTool.send behavior and callback dispatch before deploying to production.
功能分析
Type: OpenClaw Skill Name: prompt-buttons Version: 1.0.2 The prompt-buttons skill is a legitimate utility designed to standardize interactive button menus for Telegram agents. The code in index.js and lib/sendWithButtons.js implements basic validation (e.g., limiting button labels to 3 characters) and provides helper functions for Yes/No and numbered menus. The SKILL.md documentation correctly outlines security best practices, such as avoiding secrets in callback data, and the overall behavior aligns perfectly with the stated purpose of improving user experience.
能力评估
Purpose & Capability
The skill's name/description (short tappable button menus) aligns with the code and SKILL.md: helper functions to build prompts, validation, and a send helper that calls the agent message tool. No unrelated binaries, env vars, or external endpoints are requested. Minor inconsistency: SKILL.md permits button labels up to 10 chars in some cases, but lib/sendWithButtons enforces a 1–3 char limit (throws on >3). This is a documentation vs implementation mismatch but not a security problem.
Instruction Scope
SKILL.md stays largely on-topic (how to construct buttons, callback_data handling, fallback rendering). It also instructs skill authors to register callback handlers, acknowledge Telegram callback_query, and spawn background subagents for long-running tasks — these are reasonable for the stated use. SKILL.md includes install/discovery instructions that reference writing files under /data/workspace and enabling an entry in /data/.openclaw/openclaw.json; those are operational installation steps (not executed automatically) but are out-of-band instructions that affect agent configuration and should be followed with caution by an operator.
Install Mechanism
There is no install spec and no external downloads — the skill is instruction/code-only. This is low-risk: nothing in the package pulls code from arbitrary URLs or adds system-wide binaries.
Credentials
The skill requests no environment variables, no credentials, and no config paths. The SKILL.md explicitly warns not to include secrets in callback_data. The code likewise does not attempt to read env vars or files.
Persistence & Privilege
The skill does not request always:true and won’t be force-enabled by default. However, SKILL.md tells operators how to enable the skill in /data/.openclaw/openclaw.json and to place files under /data/workspace; these are manual installation steps that grant the skill presence in the agent environment (normal for skills) and should be applied only by a trusted operator.
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 prompt-buttons
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /prompt-buttons 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
v1.0.2
Telegram inline buttons with graceful fallback for OpenClaw agents
元数据
Slug prompt-buttons
版本 1.0.2
许可证 MIT-0
累计安装 1
当前安装数 1
历史版本数 1
常见问题

Prompt Buttons 是什么?

A comprehensive helper skill that wraps agent prompts with short, consistent tappable button menus (Yes/No, single digits, or small option sets) so users can... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 149 次。

如何安装 Prompt Buttons?

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

Prompt Buttons 是免费的吗?

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

Prompt Buttons 支持哪些平台?

Prompt Buttons 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 Prompt Buttons?

由 tanjinlimkelvin-dot(@tanjinlimkelvin-dot)开发并维护,当前版本 v1.0.2。

💬 留言讨论