OpenClaw Doctor

一键检查 OpenClaw 系统健康状态（配置、模型、cron、记忆）

Security Scan

Instruction-only health-check skill that reads local OpenClaw configuration and workspace files and runs OpenClaw CLI commands; its requirements and instructions match the described purpose.

Purpose & Capability

Name/description (system health checks) align with requested binaries (node) and the runtime steps (checking openclaw.json, running `openclaw status` / `openclaw cron list`, inspecting MEMORY.md/TOOLS.md). Nothing requested is unrelated to a diagnostics tool.

Instruction Scope

SKILL.md confines actions to reading local config/files and running OpenClaw CLI commands, which is appropriate; however it will inspect workspace files (MEMORY.md, daily notes, TOOLS.md) and may surface API keys or other secrets found in configs—these are within scope for a health check but are sensitive.

Install Mechanism

Instruction-only skill with no install spec or code files; no downloads or third-party packages are installed. Lowest-risk install footprint.

Credentials

No environment variables or credentials are declared or required. That said, the instructions talk about checking API key validity and channel connections, which implies reading whatever keys/tokens are stored in local config files; this is proportionate to diagnostics but could expose secrets if present in those files.

Persistence & Privilege

always is false, no install hooks or config writes are declared, and the skill is user-invocable. No elevated or persistent privileges are requested.

Assessment

This skill is an instruction-only diagnostics tool and appears coherent with its stated purpose. Before installing or running it, review your openclaw.json and workspace files (MEMORY.md, TOOLS.md, daily notes) because the agent will read those to produce the report and may reveal API keys or other sensitive content found there. Only run in a trusted environment and avoid enabling the skill if you do not want automated reads of your local/OpenClaw workspace. If you need stricter privacy, remove or rotate secrets from config files beforehand or run the checks manually.

🩺 OpenClaw 健康检查

一键全面检查 OpenClaw 系统状态，发现问题并给出修复建议。

检查项目

1. 系统状态

• OpenClaw 版本
• Node.js 版本
• 运行时间、内存占用
• Gateway 进程状态

2. 配置检查

• 配置文件是否有效（openclaw.json）
• 频道连接状态（飞书/Telegram 等）
• 模型配置是否正确
• 权限设置是否合理

3. 模型连通性

• 主模型能否正常响应
• Fallback 模型是否可用
• API Key 是否有效

4. 定时任务

• Cron 任务数量和状态
• 最近一次执行是否成功
• 有没有连续失败的任务

5. 记忆系统

• MEMORY.md 是否存在
• daily notes 最近更新时间
• TOOLS.md 是否有内容

6. 技能加载

• 已安装技能数量
• 技能格式是否正确
• 有没有重复或冲突的技能

使用方式

AI 调用步骤：

1. 检查 openclaw.json 配置文件
2. 执行 openclaw status 获取系统状态
3. 执行 openclaw cron list 查看定时任务
4. 检查 workspace 文件（MEMORY.md, TOOLS.md 等）
5. 汇总结果，给出健康评分（0-100）和修复建议

输出格式

🩺 OpenClaw 健康检查报告

系统状态：✅/⚠️/❌
- 版本：xxx
- 运行时间：xxx
- 内存：xxx

配置：✅/⚠️/❌
- 频道：xxx
- 模型：xxx

定时任务：✅/⚠️/❌
- 总数：x
- 正常：x
- 异常：x

记忆系统：✅/⚠️/❌

健康评分：xx/100
建议修复：xxx

注意事项

• 检查时不要修改任何配置
• 发现问题先报告，不要自动修复（除非用户明确要求）
• 敏感信息（API Key 等）用 *** 脱敏