← 返回 Skills 市场

ClawGuard Scanner

Name: ClawGuard Scanner
Author: frrrrrrrrank

作者 Frrrrrrrrank · GitHub ↗ · v0.1.0 · MIT-0

cross-platform ⚠ suspicious

275

总下载

当前安装

版本数

在 OpenClaw 中安装

/install clawguard-scanner

功能描述

Security scanner for OpenClaw skills. Run before installing any skill to detect prompt injection, data exfiltration, permission overreach, suspicious URLs, d...

使用说明 (SKILL.md)

ClawGuard — OpenClaw Skill Security Scanner

You are a security-conscious assistant. Before the user installs or uses any third-party OpenClaw skill, you MUST run a security scan using ClawGuard.

When to use this skill

The user asks to install a new skill (e.g., clawhub install \x3Cskill-name>)
The user asks you to review or audit a skill for safety
The user asks you to check if a skill is safe to use
The user points you at a skill directory or SKILL.md file

How to scan

Run the following command on the skill directory:

npx clawguard scan \x3Cpath-to-skill-directory>

For JSON output (useful for programmatic analysis):

npx clawguard scan \x3Cpath-to-skill-directory> --json

To check only specific rules:

npx clawguard scan \x3Cpath-to-skill-directory> --rules prompt-injection,data-exfiltration

Interpreting results

ClawGuard checks for 6 types of security issues:

Severity	Rules
CRITICAL	`prompt-injection` — instruction overrides, role switching, hidden payloads
CRITICAL	`data-exfiltration` — reading sensitive files (~/.ssh, ~/.aws) and sending externally
HIGH	`permission-overreach` — requesting sudo, rm, docker, or excessive env vars
HIGH	`suspicious-urls` — IP-based URLs, URL shorteners, known malicious domains
HIGH	`dangerous-commands` — rm -rf /, curl \| sh, system file modification
MEDIUM	`metadata-mismatch` — undeclared env vars, unused declared binaries

How to respond to scan results

If the scan PASSES (exit code 0, no findings):

Tell the user the skill passed all security checks and is safe to install. Proceed with the installation.

If the scan FAILS (exit code 1, findings detected):

Show the user ALL findings clearly, grouped by severity
For CRITICAL findings: Strongly recommend NOT installing the skill. Explain the specific risk.
For HIGH findings: Warn the user and ask for explicit confirmation before proceeding
For MEDIUM findings: Inform the user but allow installation if they acknowledge the warnings
Never silently skip or hide any finding

Example interaction flow:

User: "Install the cool-scraper skill"

You should:

First locate the skill directory
Run npx clawguard scan \x3Cskill-dir>
Report the results to the user
Only proceed with installation if the scan passes or the user explicitly accepts the risks

Important notes

Always scan BEFORE installation, never after
If ClawGuard is not installed, run npm install -g clawguard first
If a skill contains scripts (.sh, .py, .js), ClawGuard will scan those too
A clean scan does not guarantee absolute safety — it catches known patterns only
For skills that interact with external websites, note that content at those URLs may change over time (a safe link today could become malicious tomorrow)

安全使用建议

This skill appears to do what it says: run an npm-based scanner against skill directories. Before you run it, verify the clawguard npm package and GitHub repo (author, recent activity, package name/version), prefer running it in an isolated environment (container or VM), avoid 'npm install -g' unless you trust the publisher, and consider using 'npx --package clawguard@<version>' or inspecting the package tarball before executing to pin a trusted version. If you need higher assurance, ask for a reproducible install spec (source URL, SHA256 or a pinned version) or scan the package contents yourself before running.

功能分析

Type: OpenClaw Skill Name: clawguard-scanner Version: 0.1.0 The skill functions as a security gatekeeper, using SKILL.md instructions to mandate that the AI agent execute an external npm package ('npx clawguard') before installing any other skill. While the stated intent is defensive, this creates a high-risk behavioral override where the agent is directed to run external code as a prerequisite for standard operations, effectively intercepting the installation flow. The reliance on a third-party package (hosted at a GitHub repository with a potentially placeholder name 'Frrrrrrrrank') to perform 'critical' security checks on local directories constitutes a significant risk and permission overreach without explicit evidence of the external package's integrity.

能力评估

✓ Purpose & Capability

Name and description describe a security scanner and the only declared requirement is 'npx', which matches the instructions to run 'npx clawguard'. The requested capabilities align with the stated purpose.

ℹ Instruction Scope

SKILL.md contains clear, narrowly scoped instructions for scanning skill directories and interpreting results. It does instruct the agent to always run a scan before installing — a strong rule but consistent with the skill. It also tells the user to run 'npm install -g clawguard' if ClawGuard is not installed, which expands scope to performing system installs if the package is not present.

⚠ Install Mechanism

No install spec is provided; the skill relies on 'npx clawguard' (or optionally a global 'npm install -g clawguard'). That means running code fetched from the npm registry at runtime with no pinned version, checksum, or guidance to verify publisher/trust. For a security tool this is functionally coherent but increases risk because arbitrary remote code will execute.

✓ Credentials

The skill declares no environment variables, credentials, or config paths. SKILL.md does not request secrets or unrelated credentials. This is proportionate to being a scanner.

✓ Persistence & Privilege

The skill is instruction-only, has no install hooks, and 'always' is false. It does not request persistent privileges or modify other skills' configs in the instructions.

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 clawguard-scanner
安装完成后，直接呼叫该 Skill 的名称或使用 /clawguard-scanner 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

v0.1.0

Initial release: security scanner for OpenClaw skills with 6 detection rules

元数据

Slug clawguard-scanner

版本 0.1.0

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 1

常见问题