← 返回 Skills 市场
2600
总下载
0
收藏
9
当前安装
2
版本数
在 OpenClaw 中安装
/install unraid
功能描述
Query and monitor Unraid servers via the GraphQL API. Use when the user asks to 'check Unraid', 'monitor Unraid', 'Unraid API', 'get Unraid status', 'check disk temperatures', 'read Unraid logs', 'list Unraid shares', 'Unraid array status', 'Unraid containers', 'Unraid VMs', or mentions Unraid system monitoring, disk health, parity checks, or server status.
安全使用建议
This skill looks like a legitimate Unraid monitoring helper, but the declared metadata is incomplete and the scripts perform filesystem I/O in your home directory and expect environment variables and helper binaries. Before installing or running: 1) Inspect $HOME/.clawdbot/credentials/unraid/config.json (if it exists) to ensure it only contains servers/API keys you intend to expose to this tool. 2) Be aware the scripts save debug and report files (e.g., ${NAME}_debug.json and $HOME/clawd/memory/bank/unraid-inventory.md) — run in an isolated directory or container if you don’t want these persisted. 3) Verify you are willing to provide an Unraid API key (give the Viewer role only) and that keys stored for other services are not stored in the same config. 4) The helper uses curl -k (skips TLS verification) — consider editing scripts to remove -k or ensure you use valid TLS certs or a private network. 5) Ensure jq and curl are available on the host. If these gaps (missing metadata about env vars, binaries, and config paths) are acceptable and you trust the source, the scripts are coherent for their stated monitoring purpose; otherwise ask the publisher to correct metadata or provide a version that documents and limits file access.
功能分析
Type: OpenClaw Skill
Name: unraid
Version: 1.0.1
The skill is classified as suspicious due to two main indicators. First, the `scripts/dashboard.sh` script attempts to read configuration from `$HOME/.clawdbot/credentials/unraid/config.json` and write output to `$HOME/clawd/memory/bank/unraid-inventory.md`. While this might be intended for the agent's self-configuration and reporting, it demonstrates the capability to access and modify internal agent-specific files, which is a high-risk behavior. Second, the `scripts/unraid-query.sh` script uses `curl -k` (insecure) for network requests, disabling SSL certificate validation and making connections vulnerable to Man-in-the-Middle attacks.
能力评估
Purpose & Capability
The name, description, SKILL.md, and scripts are coherent with an Unraid monitoring skill (GraphQL queries for system, disks, containers, logs). However the registry metadata claims no required env vars or binaries while the scripts clearly depend on UNRAID_URL, UNRAID_API_KEY, curl, and jq — an inconsistency between declared requirements and actual capability.
Instruction Scope
SKILL.md instructs use of the helper scripts and to export UNRAID_URL and UNRAID_API_KEY. The included scripts (dashboard.sh, unraid-query.sh, examples) read a configuration file ($HOME/.clawdbot/credentials/unraid/config.json), call the Unraid endpoint, and write output/debug files (e.g., ${NAME}_debug.json and $HOME/clawd/memory/bank/unraid-inventory.md). The skill's runtime instructions and code therefore access and create files under the user's home and may persist query results — this scope is broader than the registry metadata indicates and should be validated.
Install Mechanism
No install spec is provided (instruction-only with bundled scripts). Nothing is downloaded or written during an install step beyond the included files, so install-time risk is low.
Credentials
Requesting an Unraid API key (Viewer role) is appropriate for a monitoring skill, but the registry lists no required env vars while the code expects UNRAID_URL and UNRAID_API_KEY. The dashboard script also reads $HOME/.clawdbot/credentials/unraid/config.json (which may contain multiple server entries and API keys) — the skill implicitly expects access to stored credentials but did not declare that in metadata. Additionally scripts expect curl and jq but metadata lists no required binaries.
Persistence & Privilege
The skill does not request forced always-on or autonomously modify other skills. However it writes files into the user's home (debug JSON and an inventory at $HOME/clawd/memory/bank/unraid-inventory.md) and reads a credentials config under $HOME/.clawdbot; this is normal for a local CLI tool but is persistent and could expose collected data if you run it in a shared environment. The skill also uses curl -k (ignoring TLS verification) which may reduce transport security.
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 unraid - 安装完成后,直接呼叫该 Skill 的名称或使用
/unraid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
v1.0.1
- Added a version field to SKILL.md header (version 1.0.1).
- No other content changes; documentation and usage remain unchanged.
v1.0.0
Initial release
元数据
常见问题
Unraid 是什么?
Query and monitor Unraid servers via the GraphQL API. Use when the user asks to 'check Unraid', 'monitor Unraid', 'Unraid API', 'get Unraid status', 'check disk temperatures', 'read Unraid logs', 'list Unraid shares', 'Unraid array status', 'Unraid containers', 'Unraid VMs', or mentions Unraid system monitoring, disk health, parity checks, or server status. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 2600 次。
如何安装 Unraid?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install unraid」即可一键安装,无需额外配置。
Unraid 是免费的吗?
是的,Unraid 完全免费(开源免费),可自由下载、安装和使用。
Unraid 支持哪些平台?
Unraid 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 Unraid?
由 jmagar(@jmagar)开发并维护,当前版本 v1.0.1。
推荐 Skills