InfoDashboard Skill
Use this as a guided, confirmation-heavy SOP. Do not compress the whole setup into one reply and do not perform state-changing actions without explicit user confirmation.
Core Rules
- - Move one phase at a time.
- Before any state-changing action, ask for confirmation.
- If local state already exists, show what you found and ask whether to keep it.
- InfoDashboard generation uses server-side provider config, not the OpenClaw agent's own model or API key.
- This skill must not rely on any request-time model or provider overrides.
- Only server-side config files (
.env.local) may control LLM provider selection. - Do not ask the user to paste API keys or database passwords into chat.
- Prefer guiding the user to edit
.env.local themselves. - Do not offer to write secrets into config files on the user's behalf.
- Once setup is complete and the user clearly asks to generate a dashboard, do not ask for a second confirmation before submitting the generation request.
Optional Skill Config
If present, read defaults from ~/.openclaw/openclaw.json under:
CODEBLOCK0
Use repoDir and url only as defaults. Still confirm before acting.
SOP Phases
1. Clone Or Reuse Existing Repo
Load references/clone.md.
Establish which InfoDashboard checkout to use. If a checkout already exists, show the path and ask whether to reuse it.
2. Configure Environment
Load references/config.md.
Configure .env.local with LLM provider keys and database connection details. This includes verifying that Docker and the frpc SOCKS5 tunnel are in place.
3. Start And Verify
Load references/startup.md.
Start the FastAPI server and confirm it is healthy at GET {url}/.
4. Generate A Dashboard
Load references/generate-flow.md.
Submit the user's natural-language requirement and follow the SSE stream until a dashboard URL is returned.
Response Style
- - Keep each step short and explicit.
- Prefer 2-3 concrete options when the user must choose.
- Always include the recommended option first and explain why in one sentence.
- After a step completes, say what changed and what the next confirmation is for.
- When returning a dashboard URL, place the raw absolute URL on its own line with no bold, markdown link syntax, code formatting, or tables.
InfoDashboard 技能
请将此作为一份引导式、需多次确认的标准操作流程。不要将整个设置压缩到一次回复中,未经用户明确确认不得执行状态变更操作。
核心规则
- - 一次只推进一个阶段。
- 在执行任何状态变更操作前,先请求确认。
- 如果本地状态已存在,展示你所发现的内容,并询问是否保留。
- InfoDashboard 的生成使用服务端提供者配置,而非 OpenClaw 智能体自身的模型或 API 密钥。
- 此技能不得依赖任何请求时的模型或提供者覆盖设置。
- 只有服务端配置文件(.env.local)可以控制 LLM 提供者的选择。
- 不要要求用户将 API 密钥或数据库密码粘贴到聊天中。
- 优先引导用户自行编辑 .env.local 文件。
- 不要主动替用户将机密信息写入配置文件。
- 一旦设置完成且用户明确要求生成仪表盘,在提交生成请求前不要再次请求确认。
可选技能配置
如果存在,从 ~/.openclaw/openclaw.json 中读取默认值,路径如下:
jsonc
{
skills: {
entries: {
infodashboard: {
enabled: true,
config: {
repoDir: /path/to/InfoDashboard,
url: http://localhost:8001
}
}
}
}
}
仅将 repoDir 和 url 作为默认值使用。执行操作前仍需确认。
标准操作流程阶段
1. 克隆或复用现有仓库
加载 references/clone.md。
确定要使用的 InfoDashboard 检出副本。如果已存在检出副本,显示其路径并询问是否复用。
2. 配置环境
加载 references/config.md。
使用 LLM 提供者密钥和数据库连接详情配置 .env.local 文件。这包括验证 Docker 和 frpc SOCKS5 隧道是否就位。
3. 启动并验证
加载 references/startup.md。
启动 FastAPI 服务器,并通过 GET {url}/ 确认其运行正常。
4. 生成仪表盘
加载 references/generate-flow.md。
提交用户的自然语言需求,并跟随 SSE 数据流,直到返回仪表盘 URL。
回复风格
- - 保持每一步简短明确。
- 当用户需要选择时,优先提供 2-3 个具体选项。
- 始终将推荐选项放在首位,并用一句话说明原因。
- 完成一个步骤后,说明发生了什么变化以及下一步确认的内容。
- 返回仪表盘 URL 时,将原始绝对 URL 单独放在一行,不使用加粗、Markdown 链接语法、代码格式或表格。