## 捆绑包插件

该项目也可以作为 ClawHub 捆绑包插件发布。```bash
# Install the bundle plugin variant
openclaw plugins install clawhub:openclaw-security-monitor-bundle

## 架构```
openclaw-security-monitor/
  .codex-plugin/
    plugin.json          # Bundle-plugin manifest for ClawHub/OpenClaw plugin publishing
  scripts/
    scan.sh              # 41-point threat scanner (v5.5.0)
    remediate.sh         # Orchestrator: scan + per-check remediation
    remediate/
      _common.sh         # Shared helpers (log, confirm, fix_perms)
      check-01-c2-ips.sh ... check-41-device-identity-spoofing.sh  # 41 consolidated per-check scripts
      check-42-*.sh ... check-62-*.sh  # retained legacy advisory-specific helpers
    clawhub-scan.sh      # Scan all installed ClawHub skills against IOC database
    dashboard.sh         # CLI security dashboard with witr
    network-check.sh     # Network activity monitor
    daily-scan-cron.sh   # Cron wrapper + Telegram alerts
    telegram-setup.sh    # Telegram notification setup
    update-ioc.sh        # IOC database updater
  ioc/
    c2-ips.txt           # Known C2 IP addresses
    malicious-domains.txt # Payload/exfil domains
    file-hashes.txt      # Known malicious file hashes
    malicious-publishers.txt  # Blacklisted ClawHub accounts
    malicious-skill-patterns.txt  # Malicious skill naming patterns
  dashboard/
    server.js            # Node.js HTTP server (zero npm deps)
    index.html           # Single-file dark-themed SPA
  docs/
    threat-model.md      # Threat model and attack vectors

### 检查项 2：AMOS 窃密器 / AuthTool 标记

**严重性：** 严重

**含义：** 某个技能包含与 Atomic Stealer (AMOS) macOS 信息窃取器、NovaStealer 或 "AuthTool" 社会工程二进制文件相关的模式。

**修复措施：**```bash
# Remove the malicious skill immediately
openclaw skill remove <skill-name>

# Check if AMOS was executed (macOS)
# Look for suspicious LaunchAgents
ls ~/Library/LaunchAgents/ | grep -ivE "com\.apple|com\.openclaw\.security"

# Check for unexpected login items (macOS)
osascript -e 'tell application "System Events" to get the name of every login item'

# Check browser extensions were not tampered with
ls ~/Library/Application\ Support/Google/Chrome/Default/Extensions/

# If AMOS was executed: rotate ALL credentials (SSH, AWS, crypto wallets, browser passwords)

### Check 4: Credential Exfiltration Endpoints

**Severity:** CRITICAL

**What it means:** A skill sends data to known exfiltration services (webhook.site, pipedream.net, ngrok.io, etc.).

**Remediation:**```bash
# Remove the skill
openclaw skill remove <skill-name>

# Block exfiltration domains at the host level
# macOS/Linux:
sudo sh -c 'echo "127.0.0.1 webhook.site" >> /etc/hosts'
sudo sh -c 'echo "127.0.0.1 pipedream.net" >> /etc/hosts'
sudo sh -c 'echo "127.0.0.1 hookbin.com" >> /etc/hosts'
sudo sh -c 'echo "127.0.0.1 requestbin.com" >> /etc/hosts'

# Flush DNS cache (macOS)
sudo dscacheutil -flushcache && sudo killall -HUP mDNSResponder
# Flush DNS cache (Linux)
sudo systemd-resolve --flush-caches

### 检查6：Curl-Pipe 攻击

**严重性：** 警告

**含义：** 技能使用了 `curl | sh` 或 `wget | bash` 模式来下载并执行远程脚本。

**修复方法：**```bash
# Remove the skill
openclaw skill remove <skill-name>

# Review and restrict agent tool access
openclaw config set tools.deny '["exec","process"]'

# Block curl-pipe at the shell level (add to ~/.bashrc or ~/.zshrc)
# This creates an alias that warns on pipe-to-shell patterns:
alias curl='() { if [[ "$*" == *"|"*"sh"* ]] || [[ "$*" == *"|"*"bash"* ]]; then echo "BLOCKED: curl-pipe detected"; return 1; fi; command curl "$@"; }'

### 检查8: 技能完整性哈希

**严重性:** 警告

**含义:** 自上次扫描以来，SKILL.md文件已被修改，这可能表明存在篡改或供应链威胁。

**修复:**```bash
# Review what changed
diff ~/.openclaw/logs/skill-hashes.sha256.prev ~/.openclaw/logs/skill-hashes.sha256

# For each changed skill, inspect the diff
cd ~/.openclaw/workspace/skills/<skill-name>
git diff HEAD~1 SKILL.md   # If under version control

# If change is unexpected, reinstall from ClawHub
openclaw skill remove <skill-name>
openclaw skill install <skill-name>

# Enable skill pinning (if supported)
openclaw config set skills.autoUpdate false

### Check 10: Memory Poisoning

**严重性：** **严重（CRITICAL）**

**含义：** SOUL.md、MEMORY.md 或 IDENTITY.md 包含指令覆盖模式（例如，尝试重新定义智能体行为的角色重置语言）。

**修复建议：**```bash
# Review poisoned files
cat ~/.openclaw/workspace/SOUL.md
cat ~/.openclaw/workspace/MEMORY.md
cat ~/.openclaw/workspace/IDENTITY.md

# Remove injected content manually (edit the file, remove injected lines)
# Or restore from backup:
git checkout HEAD -- ~/.openclaw/workspace/SOUL.md

# Protect memory files from modification
chmod 444 ~/.openclaw/workspace/SOUL.md
chmod 444 ~/.openclaw/workspace/MEMORY.md
chmod 444 ~/.openclaw/workspace/IDENTITY.md

# Investigate which skill wrote to memory
grep -rl "SOUL\.md\|MEMORY\.md" ~/.openclaw/workspace/skills/

### 检查12：外部二进制下载

**严重性：** 警告

**含义：** 技能引用了可下载的二进制文件（.exe、.dmg、.pkg、带密码的.zip）或已知的恶意下载URL。

**修复建议：**```bash
# Remove the skill
openclaw skill remove <skill-name>

# Check Downloads folder for suspicious binaries (macOS)
ls -la ~/Downloads/*.{exe,dmg,pkg,zip,msi} 2>/dev/null

# Verify Gatekeeper is active (macOS)
spctl --status  # should say "assessments enabled"

# Check if any unsigned apps were installed (macOS)
sudo find /Applications -name "*.app" -exec codesign -v {} \; 2>&1 | grep "invalid"

# Linux: check /tmp and /var/tmp for suspicious downloads
find /tmp /var/tmp -name "*.sh" -o -name "*.elf" -o -name "*.bin" -mtime -7 2>/dev/null

### 检查项14：WebSocket安全 (CVE-2026-25253)

**严重级别:** 严重

**含义说明:** 网关WebSocket接受来自任意来源的连接，允许通过恶意网页实现一键远程代码执行(RCE)。

**修复建议:**```bash
# Update OpenClaw to latest version (patch included in 2026.2.3+)
openclaw update

# Verify the fix
curl -s -o /dev/null -w "%{http_code}" \
  -H "Connection: Upgrade" -H "Upgrade: websocket" \
  -H "Sec-WebSocket-Key: dGhlIHNhbXBsZSBub25jZQ==" \
  -H "Sec-WebSocket-Version: 13" \
  -H "Origin: http://evil.attacker.com" \
  http://127.0.0.1:18789/
# Should return 403 or 401, NOT 101

# If update is not available, restrict gateway to localhost
openclaw config set gateway.bind localhost

# Add firewall rule (macOS)
sudo /usr/libexec/ApplicationFirewall/socketfilterfw --add /usr/local/bin/node
# Linux (iptables)
sudo iptables -A INPUT -p tcp --dport 18789 -s 127.0.0.1 -j ACCEPT
sudo iptables -A INPUT -p tcp --dport 18789 -j DROP

### 检查项 16：敏感环境信息泄露

**严重性：** WARNING / CRITICAL (如果发现硬编码的API密钥)

**含义：** 技能读取了敏感文件 (.env, .ssh, .aws/credentials) 或包含硬编码的API密钥 / Moltbook令牌。

**修复措施：**```bash
# Remove the skill
openclaw skill remove <skill-name>

# Rotate exposed credentials immediately
# OpenAI:
# Go to https://platform.openai.com/api-keys and regenerate
# Anthropic:
# Go to https://console.anthropic.com/settings/keys and regenerate

# Move secrets to a secrets manager instead of .env files
# macOS: Use Keychain Access
security add-generic-password -a "$USER" -s "OPENAI_API_KEY" -w "<new-key>"

# Linux: Use pass or secret-tool
secret-tool store --label="OPENAI_API_KEY" service openai key api

# Restrict skill filesystem access
openclaw config set sandbox.mode all

### 检查18：工具策略 / 高级工具审计

**严重性:** 严重

**含义:** 高级工具启用了通配符访问权限，或者工具拒绝列表为空 — 允许代理执行任意命令。

**修复方法:**```bash
# Restrict elevated tools to specific principals
openclaw config set tools.elevated.allowFrom '["your-user-id"]'

# Add dangerous tools to the deny list
openclaw config set tools.deny '["exec","process","browser","filesystem-write"]'

# Require approval for elevated actions
openclaw config set tools.elevated.requireApproval true

# Review current tool permissions
openclaw config get tools

### 检查 20：mDNS/Bonjour 暴露

**严重性：** WARNING

**含义：** mDNS 正在以 "full" 模式广播，向局域网通告网关的存在、路径和 SSH 端口。

**修复措施：**```bash
# Disable mDNS broadcasting
openclaw config set discovery.mdns.mode off

# Or set to minimal (name only, no paths)
openclaw config set discovery.mdns.mode minimal

# Verify mDNS is not broadcasting (macOS)
dns-sd -B _openclaw._tcp local
# Should return no results

# Disable Bonjour for the gateway specifically (macOS)
sudo defaults write /Library/Preferences/com.apple.mDNSResponder.plist NoMulticastAdvertisements -bool YES

### Check 22: Persistence Mechanisms

**Severity:** WARNING

**What it means:** There are LaunchAgents, cron entries, or systemd services referencing OpenClaw that are not the known security monitor.

**Remediation:**```bash
# macOS: Review LaunchAgents
ls -la ~/Library/LaunchAgents/ | grep -iE "openclaw|clawdbot|moltbot"
# Remove unauthorized agents:
launchctl unload ~/Library/LaunchAgents/<suspicious-plist>
rm ~/Library/LaunchAgents/<suspicious-plist>

# Check system-level LaunchDaemons (requires root)
sudo ls /Library/LaunchDaemons/ | grep -iE "openclaw|clawdbot"

# Linux: Review systemd services
systemctl --user list-units --type=service | grep -iE "openclaw|clawdbot"
# Disable unauthorized services:
systemctl --user disable --now <service-name>

# Review crontab
crontab -l | grep -iE "openclaw|clawdbot|moltbot"
# Edit and remove suspicious entries:
crontab -e

### Check 24: 日志脱敏审计

**严重性：** 警告

**含义：** 日志脱敏已禁用（敏感数据如API密钥和密码可能以明文形式出现在日志中），或者日志目录为全局可读。

**补救措施：**```bash
# Enable log redaction
openclaw config set logging.redactSensitive true

# Fix log directory permissions
chmod 700 ~/.openclaw/logs
chmod 700 /tmp/openclaw 2>/dev/null

# Verify logs don't contain sensitive data
grep -iE "sk-[a-zA-Z0-9]{10}|password|token.*=" ~/.openclaw/logs/*.log

# Set up log rotation (macOS - newsyslog)
sudo sh -c 'echo "$HOME/.openclaw/logs/*.log 640 7 1000 * J" >> /etc/newsyslog.conf'
# Linux - logrotate
cat > /etc/logrotate.d/openclaw << 'LOGROTATE'
/home/*/.openclaw/logs/*.log {
    weekly
    rotate 4
    compress
    missingok
    notifempty
    create 0600 root root
}
LOGROTATE

### 检查 26：Exec-Approvals 配置

**严重级别：** CRITICAL

**含义：** `exec-approvals.json` 允许在未确认的情况下远程执行命令。CVE-2026-25253 漏洞利用链使用了 `exec.approvals.set` 来禁用确认提示。

**修复方法：**```bash
# Set exec approvals to deny for non-owner nodes
# Edit ~/.openclaw/exec-approvals.json manually:
# Set "security": "deny" for all non-owner nodes
# Set "ask": "always"

# Fix permissions
chmod 600 ~/.openclaw/exec-approvals.json

# Verify
cat ~/.openclaw/exec-approvals.json | python3 -m json.tool

### 检查 28：Node.js 版本 / CVE-2026-21636

**严重性：** 警告

**含义：** Node.js 版本低于 22.12.0，且存在 CVE-2026-21636 漏洞（通过 Unix 域套接字绕过权限模型），该漏洞可导致沙箱逃逸。

**修复建议：**```bash
# macOS
brew install node@22
# Or update existing
brew upgrade node

# Linux (NodeSource)
curl -fsSL https://deb.nodesource.com/setup_22.x | sudo bash -
sudo apt-get install -y nodejs

# Verify
node --version  # Should be >= v22.12.0

### 检查 30：VS Code 扩展木马检测

**严重性：** 严重

**含义：** 已发现虚假的 "ClawdBot Agent" VS Code 扩展会安装 ScreenConnect RAT 及远程访问木马。OpenClaw 没有官方的 VS Code 扩展。

**修复方法：**```bash
# Remove any clawdbot/openclaw VS Code extensions immediately
rm -rf ~/.vscode/extensions/*clawdbot*
rm -rf ~/.vscode/extensions/*openclaw*
rm -rf ~/.vscode/extensions/*moltbot*

# Check for ScreenConnect/ConnectWise artifacts
# macOS:
find /Applications -name "*ScreenConnect*" -o -name "*ConnectWise*"
# Linux:
find /opt -name "*screenconnect*" -o -name "*connectwise*"

# Scan for persistence left by the trojan
ls ~/Library/LaunchAgents/ | grep -ivE "com\.apple"

### 检查项 32：MCP 服务器安全性

**严重性：** 严重

**含义：** MCP 服务器配置了无限制访问（`enableAllProjectMcpServers=true`）或在工具描述中包含提示注入模式——从而可能导致工具投毒和拉地毯攻击。

**修复措施：**```bash
# Use explicit MCP server allowlist
openclaw config set mcp.enableAllProjectMcpServers false

# Review each MCP server's tool descriptions
cat ~/.openclaw/mcp.json | python3 -m json.tool

# Pin MCP server versions (prevent rug pulls)
# In mcp.json, use specific versions, not "latest"

# Run MCP servers in separate containers
# Use Cisco MCP Scanner for analysis:
# https://github.com/cisco-ai-defense/mcp-scanner

# Audit MCP server network access
# Check which external endpoints each server contacts
lsof -i -nP | grep -E "node|mcp" | grep ESTABLISHED

### Check 34: SSRF 防护 (CVE-2026-26322, CVE-2026-27488)

**严重性：** 警告

**含义：** 网关工具端点或 cron webhook 处理器存在服务器端请求伪造漏洞，允许攻击者访问内部服务。

**修复方法：**```bash
# Update to v2026.2.10+ for SSRF patches
openclaw update

# Restrict outbound network access
openclaw config set network.allowedHosts '["api.anthropic.com","github.com"]'

# Block metadata endpoints (cloud environments)
sudo iptables -A OUTPUT -d 169.254.169.254 -j DROP  # AWS/GCP metadata

### Check 36: ACP自动审批绕过 (GHSA-7jx5)

**严重性：** 警告

**含义：** 不受信任的SKILL.md元数据可以绕过交互式工具审批提示，允许技能在未经用户同意的情况下执行工具。

**补救措施：**```bash
# Update to v2026.2.12+ for the fix
openclaw update

# Disable auto-approval of ACP permissions
openclaw config set acp.autoApprove false

# Review which skills have ACP permissions
grep -rl "acp" ~/.openclaw/workspace/skills/*/SKILL.md

### 检查项 38：技能环境变量覆盖注入 (GHSA-82g8-464f-2mv7)

**严重程度：** 警告

**含义说明：** 技能环境变量覆盖可能导致 OpenClaw 流量重定向到攻击者控制的主机。

**修复建议：**```bash
# Update to v2026.2.12+ for the fix
openclaw update

# Audit skill env overrides
grep -rn "env:" ~/.openclaw/workspace/skills/*/SKILL.md

# Remove skills with suspicious env overrides
openclaw skill remove <skill-name>

### 检查40：日志投毒/WebSocket标头注入

**严重性：警告**

**含义：** WebSocket Origin/User-Agent 标头可以注入恶意内容到代理日志中，从而在代理审查日志时实现间接提示注入。

**修复措施：**```bash
# Update to latest version (includes log sanitization)
openclaw update

# Enable log redaction
openclaw config set logging.redactSensitive true

# Sanitize existing logs
sed -i.bak 's/\x1b\[[0-9;]*m//g' ~/.openclaw/logs/*.log

# Restrict log file permissions
chmod 600 ~/.openclaw/logs/*.log

### 检查42：浏览器控制路径遍历（CVE-2026-28462）

**严重程度：** 严重

**含义：** 浏览器控制跟踪/下载API中的路径遍历允许在预期目录之外进行任意文件写入。

**修复措施：**```bash
# Update to v2026.2.12+ for the fix
openclaw update

# Check for suspicious files written by browser control
find /tmp -name "*.trace" -newer /tmp -mtime -1 2>/dev/null

### Check 44: Approval Field Injection (CVE-2026-28466)

**严重性：** 严重

**含义：** 审批提示中的未清理字段可能注入内容，诱骗用户批准恶意操作。

**修复建议：**```bash
# Update to v2026.2.14+ for the fix
openclaw update

# Always carefully read approval prompts before accepting
# Enable verbose approval mode
openclaw config set approvals.verbose true

### Check 46: Webhook 拒绝服务 (CVE-2026-28478)

**严重性：** 警告

**含义：** Webhook 处理器缓冲请求体时未设置字节或时间限制，导致过大的或缓慢的请求可能耗尽内存。

**修复措施：**```bash
# Update to v2026.2.13+ for the fix
openclaw update

# Add reverse proxy with body size limits (nginx)
# client_max_body_size 1m;

### Check 48: fetchWithGuard Memory DoS (CVE-2026-29609)

**Severity:** WARNING

**What it means:** fetchWithGuard allocates entire response payloads in memory before enforcing maxBytes, causing memory exhaustion.

**Remediation:**```bash
# Update to v2026.2.14+ for the fix
openclaw update

### 检查 50：通过PATH的命令劫持 (CVE-2026-29610)

**严重性：** 严重

**含义：** OpenClaw 在解析命令时使用 PATH 但不固定为绝对路径，从而允许被植入的二进制文件拦截 exec 调用。

**补救措施：**```bash
# Update to v2026.2.14+ for the fix
openclaw update

# Remove writable directories before system dirs in PATH
# See Check 37 for detailed PATH hardening steps

### 检查 52: Google Chat Webhook 绕过 (CVE-2026-28469, CVSS 9.8)

**严重性:** 严重

**含义:** Google Chat Webhook 处理器使用首次匹配语义，允许跨账户攻击者注册匹配路径并拦截/注入消息。

**修复方法:**```bash
# Update to v2026.2.14+ for the fix
openclaw update

# Disable Google Chat integration if not actively used
openclaw config set integrations.googlechat.enabled false

# Audit webhook paths for duplicates
openclaw config get integrations.googlechat

### Check 54: Cross-Site WebSocket Hijacking (CVE-2026-32302)

**严重性：** 严重

**含义：** 在可信代理模式下，源验证绕过允许攻击者来源页面建立特权操作员WebSocket会话。

**修复措施：**```bash
# Update to v2026.3.11+ for the fix
openclaw update

# Disable trusted-proxy mode if not required
openclaw config set gateway.trustedProxy false

# Verify origin validation is active
curl -s -o /dev/null -w "%{http_code}" \
  -H "Origin: http://evil.attacker.com" \
  -H "Connection: Upgrade" -H "Upgrade: websocket" \
  http://127.0.0.1:18789/
# Should return 403, NOT 101

### 检查56：操作员权限提升 (GHSA-vmhq)

**严重性：**严重

**含义：** 拥有 operator.write 权限的账户可以访问仅管理员端点的创建/删除浏览器配置文件。

**修复措施：**```bash
# Update to v2026.3.12+ for the fix
openclaw update

# Audit operator accounts
openclaw users list --role operator

# Remove unnecessary write permissions
openclaw users update <user> --role operator.read

### 检查项 58：SANDWORM_MODE MCP 蠕虫检测

**严重性：** 严重

**含义：** 您的系统中包含 SANDWORM_MODE npm 蠕虫的痕迹——19 个钓鱼式命名的包，这些包会将恶意 MCP 服务器注入 AI 工具配置并窃取凭证。

**补救措施：**```bash
# Remove rogue MCP entries from config files
# Check: ~/.claude.json, ~/.cursor/mcp.json, ~/.continue/config.json, ~/.windsurf/mcp.json

# Uninstall malicious npm packages
npm uninstall -g @anthropic/sdk-extra claude-code-utils claude-mcp-helper

# CRITICAL: Rotate ALL credentials
# - SSH keys: ssh-keygen (regenerate and update authorized_keys)
# - AWS keys: aws iam create-access-key / delete-access-key
# - npm tokens: npm token revoke / npm token create
# - LLM API keys: regenerate on each provider dashboard

# Audit git repos for unauthorized commits
git log --since="2 weeks ago" --oneline

# Check for unauthorized SSH keys
cat ~/.ssh/authorized_keys

## Auto-Remediation

The `remediate.sh` orchestrator runs `scan.sh`, parses the results, skips CLEAN checks, and executes per-check remediation scripts for each WARNING/CRITICAL finding.```bash
# Scan + remediate (interactive)
./scripts/remediate.sh

# Auto-approve all fixes (explicit opt-in)
OPENCLAW_ALLOW_UNATTENDED_REMEDIATE=1 ./scripts/remediate.sh --yes

# Dry run — preview what would be fixed
./scripts/remediate.sh --dry-run

# Remediate a single check
./scripts/remediate.sh --check 7 --dry-run

# Run all 41 scripts without scanning
./scripts/remediate.sh --all

## 网页仪表盘

零依赖的 Node.js 服务器，运行在 18800 端口，提供以下功能：
- 扫描摘要环形图
- 32 张按颜色编码的安全检查卡片
- 网关状态与配置审计
- 通过 `witr` 显示的进程树，展示继承链
- 网络连接与监听端口
- 文件权限与提供商认证状态
- 扫描历史时间线（最近 30 次扫描）
- 每 30 秒自动刷新 + 按需扫描按钮

### 截图

**仪表盘概览** — 包含扫描摘要环形图、网关状态和 cron 状态的头部：

![仪表盘头部](https://assets.kitploit.com/production/public/readmes/22207/858043e97af92d4edd18e061d8408a6057493c1e8fa012947bc923c375d831cc.png)

**安全检查网格** — 所有 32 项检查，带按颜色编码的状态卡片（绿色=正常，黄色=警告，红色=严重）：

![安全检查](https://assets.kitploit.com/production/public/readmes/22207/01129a734eb89516ce75f9246c7e3b4b7e4296ed2588de683cb4a3573d649a54.png)

**完整仪表盘** — 包含进程树、网络连接和扫描历史的完整视图：

![完整仪表盘](https://assets.kitploit.com/production/public/readmes/22207/f3a82bb380b402cb23481af426ca010636388f02be3ec0ba53fc7ed29c972533.png)

**进程树、网络连接和监听端口：**

![进程树与网络](https://assets.kitploit.com/production/public/readmes/22207/d55b3d70b554e3227ca8582abf5337285ca7d1ecbfbf1f3afb062d7a1ae0ad9e.png)

**提供商认证、被阻止命令和扫描历史时间线：**

![提供商认证与历史](https://assets.kitploit.com/production/public/readmes/22207/5ca21c1e7c02eb2e2048794b527c5eca1e9d506e5d7440588a5117a386d5756f.png)

### API 端点

| 路由 | 方法 | 描述 |
|-------|--------|-------------|
| `/` | GET | 仪表盘界面 |
| `/api/scan` | GET | 最新扫描结果（从日志解析） |
| `/api/logs/scan` | GET | 扫描日志历史（最近 50 条） |
| `/api/logs/cron` | GET | Cron 日志条目 |
| `/api/ioc` | GET | IOC 数据库统计（每个文件的计数） |
| `/api/skills` | GET | 已安装技能列表及版本 |
| `/api/config` | GET | 配置审计（只读） |
| `/api/status` | GET | 服务器运行时间、上次扫描、IOC 统计 |
| `/api/help` | GET | CLI 命令参考 |

### LaunchAgent（macOS）

`docs/` 中的 plist 模板使用 `__HOME__` 占位符。安装方式：```bash
# Generate plist with your home directory and install
sed "s|__HOME__|$HOME|g" docs/com.openclaw.security-dashboard.plist \
  > ~/Library/LaunchAgents/com.openclaw.security-dashboard.plist

# Load the service
launchctl load ~/Library/LaunchAgents/com.openclaw.security-dashboard.plist

# Verify
curl -s http://localhost:18800/api/status | python3 -m json.tool

代理会通过 `~/.openclaw/workspace/skills/` 目录下 SKILL.md 文件中的 frontmatter 自动发现技能。安装完成后，代理将提供以下命令可用：

| 命令 | 描述 |
|------|------|
| `/security-scan` | 运行 41 项安全扫描 |
| `/security-remediate` | 自动修复常见问题 |
| `/security-dashboard` | CLI 安全仪表盘 |
| `/security-network` | 网络连接监控 |
| `/security-setup-telegram` | 配置 Telegram 告警 |

验证安装：```bash
# Check skill is visible
openclaw skills list | grep openclaw-security-monitor

# Test scan through the agent
# In the agent, type: /security-scan

### 安装先决条件（Linux）```bash
# Node.js (via NodeSource or your package manager)
curl -fsSL https://deb.nodesource.com/setup_22.x | sudo bash -
sudo apt-get install -y nodejs

# witr (check https://github.com/pranshuparmar/witr for Linux install)