Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

Cli Vscode

v1.0.0

Command-line interface for VSCode to open files, install/list extensions, manage workspaces, and check editor status with JSON output support.

0· 110·1 current·1 all-time

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for kaising-openclaw1/cli-vscode.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "Cli Vscode" (kaising-openclaw1/cli-vscode) from ClawHub.
Skill page: https://clawhub.ai/kaising-openclaw1/cli-vscode
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install cli-vscode

ClawHub CLI

Package manager switcher

npx clawhub@latest install cli-vscode
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Suspicious
high confidence
Purpose & Capability
The name/description match the included code: the tool wraps the local 'code' CLI to open files, list/install extensions, manage workspaces, and report status. However the registry metadata declared no required binaries while the SKILL.md and the code clearly require the 'code' executable in PATH — this is a minor metadata mismatch. There is also an inconsistency in licensing/monetization: _meta.json lists a commercial license and price while SKILL.md/README mention MIT (personal use) and paid licensing, which is an administrative inconsistency to clarify with the author.
Instruction Scope
SKILL.md instructs the agent only to run the included CLI wrapper and ensure the VSCode 'code' CLI is available. The runtime instructions do not ask the agent to read unrelated files or environment variables. The code does accept file/folder/extension-id arguments and will invoke the local 'code' command accordingly (expected for the stated purpose).
Install Mechanism
There is no install spec and no external downloads; the skill is instruction-only plus a single Python script included in the bundle. That keeps install risk low — nothing external is fetched or extracted by the skill itself.
Credentials
The skill requests no environment variables or credentials and the code does not read secrets or other system config. This is proportional to the described functionality.
Persistence & Privilege
always:false and no special persistence requested. However, like any skill that the agent can invoke autonomously, it can be used to open files and to install extensions on the host if the agent issues those commands. Installing arbitrary VSCode extensions is a powerful capability (it can modify the editor environment and run extension code), so consider agent autonomy policies before enabling this skill.
What to consider before installing
What to consider before installing: - Functionality: The skill is a thin wrapper around your local 'code' CLI — it will open files, add folders, list and install extensions, and report VSCode version. This matches its description. - Metadata mismatch: The package metadata does not declare the 'code' binary requirement (but both SKILL.md and the script require it) and license/price information is inconsistent. Ask the author to clarify license and the correct requirements. - Extension installs: The skill can install arbitrary VSCode extensions when invoked. If you allow the agent to call the skill autonomously, it could install extensions without further approval — extensions can execute code and could be a risk. Only enable this skill for agents you trust, or restrict/remove the 'install-extension' command if you want to limit risk. - Local impact: The tool performs local operations only (no network calls coded into the skill). Still, opening sensitive local files or adding folders is possible — avoid granting the agent broad autonomous privileges if you have sensitive data on the machine. - Code review: The included Python is small and readable; subprocess.run is invoked with argument lists (not shell=True), which reduces command-injection risk from untrusted inputs, but you should still review how your agent supplies file/ID/folder parameters. Recommended actions: confirm license and source, limit autonomous invocation or disable the install-extension command unless needed, and run in a restricted environment if you have sensitive data.

Like a lobster shell, security has layers — review code before you run it.

latestvk972n6dc8wd3qhnq80v7s9f6rx840h7g
110downloads
0stars
1versions
Updated 3w ago
v1.0.0
MIT-0

CLI-VSCode SKILL.md

Version: 1.0.0
Type: CLI Tool
Interface: Command Line + JSON


Description

CLI-VSCode 是 VSCode 的命令行接口,让 AI Agent 可以直接操作 VSCode。

支持功能:

  • 打开文件/VSCode
  • 安装/列出扩展
  • 管理工作区
  • 状态检查

Installation

确保 VSCode 已安装并且 code 命令行工具可用。

macOS

在 VSCode 中按 Cmd+Shift+P,输入 "Shell Command: Install 'code' command in PATH"

Windows

VSCode 安装时勾选 "Add to PATH"

Linux

sudo ln -s /usr/share/code/bin/code /usr/local/bin/code

Commands

# 打开文件
python cli-vscode.py open ./src/main.py

# 安装扩展
python cli-vscode.py install-extension --id esbenp.prettier-vscode

# 列出扩展
python cli-vscode.py list-extensions

# 添加到工作区
python cli-vscode.py add-folder ./tests

# 检查状态
python cli-vscode.py status

# JSON 输出 (Agent 使用)
python cli-vscode.py --json list-extensions

JSON Schema

List Extensions Response

{
  "extensions": [
    "esbenp.prettier-vscode",
    "ms-python.python",
    "GitHub.copilot"
  ]
}

Status Response

{
  "installed": true,
  "version": "1.88.0"
}

Agent Integration

OpenClaw

skill: cli-vscode
type: cli
commands:
  - open
  - install-extension
  - list-extensions
  - add-folder
  - status

Limitations

  • 需要 VSCode 已安装
  • 需要 code 命令行工具在 PATH 中
  • 不支持编辑文件内容(仅打开)

License

MIT License (个人使用)
商业许可需单独购买

Comments

Loading comments...