← 返回 Skills 市场
ClawSec
作者
Paperknight
· GitHub ↗
· v1.0.0
16801
总下载
13
收藏
261
当前安装
1
版本数
在 OpenClaw 中安装
/install clawsec
功能描述
Manage and operate ClawSec Monitor v3.0, a MITM HTTP/HTTPS proxy that logs AI agent traffic, detects exfiltration and injection threats in real time.
安全使用建议
Install or use this only if you trust and have reviewed the actual ClawSec Monitor implementation. Prefer per-process CA settings over system-wide trust, route only the agents you intend to monitor, protect and delete /tmp/clawsec logs when finished, and remove any trusted CA or Docker volume after use.
功能分析
Type: OpenClaw Skill
Name: clawsec
Version: 1.0.0
The skill describes a security monitoring tool (ClawSec Monitor) designed to intercept and inspect AI agent HTTP/HTTPS traffic for threats like data exfiltration and command injection. The `skill.md` provides instructions for the AI agent to help users manage, operate, and troubleshoot this proxy. All commands and instructions are directly related to the stated purpose of a security monitor. There is no evidence of malicious intent, prompt injection attempts to subvert the agent, or instructions for the agent to perform unauthorized actions, exfiltrate data, or establish persistence. The privileged operations (e.g., `sudo` for CA installation) are part of the legitimate setup for a security proxy, not an exploit.
能力评估
Purpose & Capability
The stated purpose is coherent with a ClawSec traffic monitor, and MITM behavior is disclosed, but the capability exposes sensitive proxied traffic including API keys, prompts, responses, and session-bearing headers.
Instruction Scope
The operational commands are related to the stated purpose, but the skill includes system trust-store changes, proxy routing, foreground/background service operation, and troubleshooting guidance without clear consent, scope, or removal steps.
Install Mechanism
The submitted package contains only skill.md while referencing a Python monitor, Dockerfile, compose file, tests, and requirements that are not included for review; the agent may be guided to run local or externally obtained high-impact code.
Credentials
HTTPS interception and a local trusted CA are proportionate to a MITM security monitor, but system-wide CA installation is broader than necessary when per-process trust is also available.
Persistence & Privilege
The artifact discloses persistent logs under /tmp/clawsec and a Docker volume that keeps CA material across restarts, but it does not provide cleanup, revocation, or log-protection guidance.
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 clawsec - 安装完成后，直接呼叫该 Skill 的名称或使用
/clawsec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
Security proxy with HTTPS MITM for Moltbot. Monitors traffic,
catches API key exfiltration, command injection, and SSH abuse.
Built after finding 341 malicious skills. One-command install.
Production-tested: 5,152 req/s, 34/34 tests passed.
元数据
常见问题
ClawSec 是什么？
Manage and operate ClawSec Monitor v3.0, a MITM HTTP/HTTPS proxy that logs AI agent traffic, detects exfiltration and injection threats in real time. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 16801 次。
如何安装 ClawSec？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install clawsec」即可一键安装，无需额外配置。
ClawSec 是免费的吗？
是的，ClawSec 完全免费（开源免费），可自由下载、安装和使用。
ClawSec 支持哪些平台？
ClawSec 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 ClawSec？
由 Paperknight（@chrisochrisochriso-cmyk）开发并维护，当前版本 v1.0.0。
推荐 Skills