← Back to Skills Marketplace
100
Downloads
0
Stars
1
Active Installs
2
Versions
Install in OpenClaw
/install system-selfcheck
Description
系统自检：一键诊断 OpenClaw 运行环境，输出结构化报告。 检查系统资源、运行时依赖、网络连通性、API 配置、Skills 状态。 预留 MiClaw / Hermes 平台适配。
Usage Guidance
This skill appears to perform the advertised diagnostics, but take care before using its auto-fix mode. Steps to reduce risk: 1) Run the script without `--fix` first to see findings. 2) Inspect the full script source yourself (you already have it) and confirm you trust the remote endpoints it touches (github.com, api.xiaomimimo.com, clawhub.ai, npm registry). 3) Avoid `--fix` on production hosts or run it in a sandbox/container where package installs won't affect critical systems. 4) Be cautious about the npx invocation — it executes code fetched from the npm registry; if you need that check, consider replacing it with a local, pinned tool. 5) If you plan to schedule automatic runs, do not enable `--fix` in unattended cron jobs and ensure the user account used has only the permissions you intend.
Capability Analysis
Type: OpenClaw Skill
Name: system-selfcheck
Version: 1.2.0
The skill is a legitimate diagnostic tool for the OpenClaw environment, performing system resource checks, dependency verification (ffmpeg, jq), and network connectivity tests to relevant domains like github.com and clawhub.ai. While it checks for the existence of API keys and includes an auto-fix feature in 'scripts/selfcheck.sh' that can install packages via pip or apt, these actions are consistent with its stated purpose and do not exhibit signs of data exfiltration or malicious intent.
Capability Tags
Capability Assessment
Purpose & Capability
The script's checks (system, memory, disk, deps, network, OpenClaw gateway/config/skills, and presence of MIMO_API_KEY/ClawHub token) match the skill description. Minor inconsistency: the registry metadata lists no required env vars, yet the script checks MIMO_API_KEY and CLAWHUB_TOKEN and reads config paths (~/.openclaw/openclaw.json, ~/.config/clawhub/config.json). This is coherent with purpose but the metadata omission should be noted.
Instruction Scope
SKILL.md plus scripts instruct running a shell script that performs many system checks and (optionally) mutating actions. The `--fix` path installs packages and runs tools that reach out to the network. The script calls `npx clawhub@latest whoami` (executes code downloaded from npm), runs curl against external endpoints, and uses package managers (pip3, apt-get) to install software. These behaviors go beyond passive diagnostics and grant the skill the ability to modify the host and execute remote code.
Install Mechanism
There is no formal install spec (instruction-only), which is low risk by itself, but the script itself performs runtime installs: pip3 install (--break-system-packages), apt-get install -y, and npx invocation. Those are effectively dynamic installs of third-party code at runtime and are a higher-risk install mechanism because they pull and execute remote code and may require elevated privileges.
Credentials
The script only checks for presence of MIMO_API_KEY and CLAWHUB_TOKEN (and does not print their values), which is proportionate to a platform health check. However, these env vars were not declared in the skill metadata. Additionally, invoking `npx clawhub@latest whoami` may interact with local auth state and could cause remote code to access credentials implicitly if present; this should be considered when deciding to run the script.
Persistence & Privilege
The skill is not force-included (always:false) and does not modify other skills' configs, but the `--fix` option performs system modifications (apt/pip installs) that typically require root and can change the runtime environment. Running with elevated privileges or scheduling the script (cron example in SKILL.md) increases risk because the skill can autonomously alter the host if invoked with `--fix` or if an operator runs it without reviewing.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install system-selfcheck - After installation, invoke the skill by name or use
/system-selfcheck - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.2.0
macOS vm_stat 内存检测; Skills 路径动态获取; OpenClaw 版本对比; --brief All clear ✅
v1.0.1
首发：系统自检 Skill，支持 OpenClaw 平台，预留 MiClaw/Hermes，三种输出模式，告警阈值
Metadata
Frequently Asked Questions
What is System Selfcheck?
系统自检：一键诊断 OpenClaw 运行环境，输出结构化报告。 检查系统资源、运行时依赖、网络连通性、API 配置、Skills 状态。 预留 MiClaw / Hermes 平台适配。 It is an AI Agent Skill for Claude Code / OpenClaw, with 100 downloads so far.
How do I install System Selfcheck?
Run "/install system-selfcheck" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is System Selfcheck free?
Yes, System Selfcheck is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does System Selfcheck support?
System Selfcheck is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created System Selfcheck?
It is built and maintained by aaroncxxx (@aaroncxxx); the current version is v1.2.0.
More Skills