扫描报告
20 /100
chatclaw
Connect your OpenClaw bot to the ChatClaw cloud dashboard for real-time remote chat, token tracking, and task management
ChatClaw is a legitimate message relay skill that connects local OpenClaw agents to a cloud dashboard via WebSocket tunnels. No malicious behavior detected; all declared capabilities match implementation.
可以安装
Approve for use. Consider pinning dependency versions in requirements.txt to reduce supply chain risk.
安全发现 1 项
| 严重性 | 安全发现 | 位置 |
|---|---|---|
| 低危 | Dependencies not version-pinned 供应链 | requirements.txt:1 |
| 资源类型 | 声明权限 | 推断权限 | 状态 | 证据 |
|---|---|---|---|---|
| 文件系统 | WRITE | WRITE | ✓ 一致 | main.py:52-80 - reads/writes openclaw.json; gateway_client.py:45-68 - reads iden… |
| 网络访问 | READ | READ | ✓ 一致 | relay_client.py:30 - connects to wss://api.sumeralabs.com; gateway_client.py:170… |
| 命令执行 | NONE | NONE | — | No subprocess, os.system, or shell command execution found |
| 环境变量 | READ | READ | ✓ 一致 | gateway_client.py:23-29 - reads OPENCLAW_DATA_DIR only; main.py:65-66 - reads CH… |
| 技能调用 | NONE | NONE | — | No skill invocation or LLM prompt manipulation detected |
| 剪贴板 | NONE | NONE | — | No clipboard access found |
| 浏览器 | NONE | NONE | — | No browser automation found |
| 数据库 | NONE | NONE | — | No database access found |
4 项发现
中危 外部 URL 外部 URL
https://app.chatclaw.com README.md:9 中危 外部 URL 外部 URL
https://opensource.org/licenses/MIT-0 README.md:89 中危 外部 URL 外部 URL
https://chatclaw.com SKILL.md:6 中危 外部 URL 外部 URL
https://app.chatclaw.com/setup SKILL.md:89 目录结构
6 文件 · 37.9 KB · 975 行 Python 3f · 707L
Markdown 2f · 266L
Text 1f · 2L
├─
gateway_client.py
Python
├─
main.py
Python
├─
README.md
Markdown
├─
relay_client.py
Python
├─
requirements.txt
Text
└─
SKILL.md
Markdown
依赖分析 3 项
| 包名 | 版本 | 来源 | 已知漏洞 | 备注 |
|---|---|---|---|---|
websockets | >=12.0 | pip | 否 | Version not pinned - allows any 12.x+ version |
cryptography | >=41.0 | pip | 否 | Version not pinned - allows any 41.x+ version |
aiohttp | >=3.9 | pip | 否 | Version not pinned - allows any 3.9+ version |
安全亮点
✓ Comprehensive SKILL.md documentation with explicit transparency notes about cloud infrastructure operators
✓ All file access scoped to $OPENCLAW_DATA_DIR with declared purpose (identity files, config, sessions)
✓ No shell execution or subprocess usage - pure asyncio/websocket architecture
✓ No credential exfiltration - API key used only for authentication to declared cloud relay
✓ No obfuscation or anti-analysis techniques detected
✓ Clean code with proper error handling and atomic file writes
✓ Sensitive files (~/.ssh, ~/.aws, .env) never accessed
✓ Ed25519 cryptography for authentication - modern and secure
✓ Auto-revert mechanism for config patches on skill disable
✓ Exponential backoff reconnection with 60s cap