← 返回 Skills 市场
158
总下载
1
收藏
1
当前安装
3
版本数
在 OpenClaw 中安装
/install qinglong-skills
功能描述
Manage QingLong (青龙) panel via REST API — cron jobs, environment variables, scripts, dependencies, subscriptions, logs and system operations. Use this skill...
安全使用建议
This skill appears to do exactly what it says: it calls your self-hosted QingLong panel using the panel URL and the client credentials you provide. Before installing or enabling it: 1) Only grant the minimum Open API scopes in QingLong (avoid full 'system' scope unless you need system-level actions). 2) Be aware that the script caches an auth token in $XDG_CACHE_HOME or $HOME/.cache/qinglong/token and that you must store the client secret somewhere (the README suggests openclaw.json); protect those files with appropriate filesystem permissions. 3) Review scripts/ql.sh locally (it’s included) so you’re comfortable with the exact API calls made. 4) If you will let an autonomous agent call skills, remember this skill can perform destructive/system operations on the panel (delete cron jobs, reset auth, run arbitrary commands) — require explicit confirmation before performing such actions. If any of those concerns are unacceptable, do not install or limit the app scopes/credentials you provide.
功能分析
Type: OpenClaw Skill
Name: qinglong-skills
Version: 1.0.2
The skill bundle is a comprehensive CLI wrapper for the QingLong panel REST API, allowing users to manage tasks, environment variables, and system settings. While the script (`scripts/ql.sh`) includes high-risk capabilities such as remote command execution and authentication resets, these are standard features of the QingLong API and are clearly documented for the user's own panel management. No evidence of malicious intent, data exfiltration to external endpoints, or obfuscation was found.
能力标签
能力评估
Purpose & Capability
Name/description match the requested resources and behavior. Required binaries (curl, jq), required env vars (QINGLONG_URL, QINGLONG_CLIENT_ID, QINGLONG_CLIENT_SECRET), included API reference and a CLI wrapper script (scripts/ql.sh) are all appropriate and expected for a QingLong panel management skill.
Instruction Scope
SKILL.md limits runtime behavior to calling scripts/ql.sh against the user's QINGLONG_URL. The instructions instruct listing before changes and confirming destructive actions, which is appropriate. Note: the skill and script expose APIs that permit system-level operations (e.g., system command-run, auth reset), so callers should grant only the scopes they actually want the skill to exercise and confirm destructive/system actions before running them.
Install Mechanism
No install spec; skill is instruction-only with a bundled shell script. No third-party downloads or archive extraction. This is low-risk from an installation perspective, though the included script will be stored and executed on the host.
Credentials
Requested environment variables (panel URL, client id, client secret) are exactly what the QingLong Open API requires. The script also uses XDG_CACHE_HOME/HOME to store a token cache, which is expected. No unrelated credentials or extraneous env vars are requested.
Persistence & Privilege
always: false (normal). The skill writes a token cache file under $XDG_CACHE_HOME or $HOME/.cache/qinglong and recommends placing credentials into openclaw.json (user-managed). Those are expected behaviors, but users should be aware that credentials and tokens will be stored on-disk and that the skill can call system-level QingLong endpoints if the configured app has those scopes.
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 qinglong-skills - 安装完成后，直接呼叫该 Skill 的名称或使用
/qinglong-skills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
v1.0.2
- Added explicit usage guidance in SKILL.md, including patterns for handling user requests and best practices before modifying resources.
- Added a "How to approach user requests" section with safety and workflow tips.
- Clarified security posture: skill only connects to user-provided, self-hosted QingLong panel.
- Moved installation and setup instructions to a new references/setup.md file and referenced it from SKILL.md.
- Improved troubleshooting section and reformatted quick reference commands for greater clarity.
- Minor metadata updates (added version, expanded trigger description) to assist with skill invocation.
v1.0.1
Refactor SKILL.md: separate OpenClaw/Claude Code setup, add ClawHub link, add npx install method
v1.0.0
Initial release: cron jobs, envs, scripts, deps, subscriptions, logs, system management. Supports OpenClaw and Claude Code CLI.
元数据
常见问题
Qinglong Skills 是什么？
Manage QingLong (青龙) panel via REST API — cron jobs, environment variables, scripts, dependencies, subscriptions, logs and system operations. Use this skill... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 158 次。
如何安装 Qinglong Skills？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install qinglong-skills」即可一键安装，无需额外配置。
Qinglong Skills 是免费的吗？
是的，Qinglong Skills 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Qinglong Skills 支持哪些平台？
Qinglong Skills 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 Qinglong Skills？
由 NNNNzs（@nnnnzs）开发并维护，当前版本 v1.0.2。
推荐 Skills