← 返回 Skills 市场
Supernal Interface CLI
作者
ianderrington
· GitHub ↗
· v1.0.0
460
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install si
功能描述
CLI tool for generating tests, scanning contracts, managing story-based tests, and setting up MCP integration in web projects.
安全使用建议
This SKILL.md describes a legitimate-seeming CLI, but before installing or running anything: (1) locate and inspect the npm package (@supernal/interface) source (GitHub/homepage, package.json, published tarball) — do not install from npm without reviewing it; (2) run any initial tests inside an isolated environment (container or disposable VM) rather than your main workstation or production repo; (3) back up your repository and avoid using auto-commit/--force flags until you’ve reviewed the changes the tool will make; (4) be cautious with commands that 'configure IDE', 'create server', or 'install agents' and verify what credentials or network connections they require; (5) if you allow an autonomous agent to run these commands, restrict its permissions and monitor its actions. If you can, ask the publisher for a homepage or source repository link and audit the package contents before trusting it.
功能分析
Type: OpenClaw Skill
Name: si
Version: 1.0.0
The `SKILL.md` describes a CLI tool with extensive, high-privilege capabilities that, while plausible for a development tool, introduce significant risk. Key indicators include `npm install -g @supernal/interface` for global package installation, `si setup-mcp` which configures IDEs and creates servers, and `si setup-claude` which installs other skills and agents. These commands grant broad system access, file system modification, and the ability to introduce further executable components, elevating the risk profile beyond benign, despite no explicit malicious intent or prompt injection attempts found within the `SKILL.md` itself.
能力评估
Purpose & Capability
The name, description, and listed commands are consistent: scanning routes/components, generating tests, story-based tests, and MCP/Claude setup are coherent with a developer CLI. However the skill metadata has no source/homepage and this SKILL.md tells the agent/user to install an external npm package (@supernal/interface) with unknown provenance — that gap in origin information is noteworthy.
Instruction Scope
The runtime instructions include commands that modify project files (init --inject, --migrate, --revert), auto-commit changes (--git-commit), create servers and configure IDEs (si setup-mcp), and install other skills/agents (si setup-claude). These are high-impact actions that go beyond read-only scanning; they can change local code, commit to git, and install software. The SKILL.md gives broad authority (e.g., 'fully automated', '--force', auto-commit flags) without describing safeguards or required confirmations.
Install Mechanism
There is no install spec in the skill bundle (instruction-only), but the documentation instructs users/agents to run 'npm install -g @supernal/interface'. Installing an unpublished/unknown npm package can execute arbitrary code. Because the package's source/homepage is missing, there's no way to inspect the package before installing from the instructions alone.
Credentials
The skill declares no required env vars or credentials, which superficially looks least-privileged. In practice many commands (git auto-commit, server setup, IDE configuration, Claude integration) will require filesystem access, network access and possibly credentials/APIs (e.g., for Claude or remote servers). The lack of declared environment requirements means those accesses are implicit and not documented.
Persistence & Privilege
The skill is not marked always:true and has no install footprint in this bundle. However the documented commands include installing software, creating servers, and installing other skills/agents (si setup-claude). If an agent executes these commands autonomously, they can create persistent software and modify system state. This combination increases blast radius if the agent is permitted to act without restrictions.
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 si - 安装完成后,直接呼叫该 Skill 的名称或使用
/si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
v1.0.0
Initial release - test generation, contract scanning, story system, MCP setup
元数据
常见问题
Supernal Interface CLI 是什么?
CLI tool for generating tests, scanning contracts, managing story-based tests, and setting up MCP integration in web projects. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 460 次。
如何安装 Supernal Interface CLI?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install si」即可一键安装,无需额外配置。
Supernal Interface CLI 是免费的吗?
是的,Supernal Interface CLI 完全免费(开源免费),可自由下载、安装和使用。
Supernal Interface CLI 支持哪些平台?
Supernal Interface CLI 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 Supernal Interface CLI?
由 ianderrington(@ianderrington)开发并维护,当前版本 v1.0.0。
推荐 Skills