npm - clawmoat - Versions diffs - 0.5.0 → 0.8.0 - Mend

clawmoat 0.5.0 → 0.8.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (44) hide show

package/CONTRIBUTING.md +4 -2
package/README.md +86 -3
package/SECURITY.md +58 -10
package/bin/clawmoat.js +298 -1
package/clawmoat-0.8.0.tgz +0 -0
package/docs/blog/386-malicious-skills.html +255 -0
package/docs/blog/40000-exposed-openclaw-instances.html +194 -0
package/docs/blog/agent-trust-protocol.html +197 -0
package/docs/blog/clawmoat-vs-llamafirewall-nemo-guardrails.html +223 -0
package/docs/blog/ibm-experts-agent-runtime-protection.html +238 -0
package/docs/blog/index.html +168 -0
package/docs/blog/mcp-30-cves-security-crisis.html +279 -0
package/docs/blog/microsoft-openclaw-workstation-security.html +234 -0
package/docs/blog/nist-ai-agent-standards-clawmoat.html +369 -0
package/docs/blog/oasis-websocket-hijack.html +205 -0
package/docs/blog/ollama-openclaw-security.html +154 -0
package/docs/blog/openclaw-enterprise-readiness-claw10.html +198 -0
package/docs/blog/openclaw-security-reckoning-2026.html +361 -0
package/docs/blog/supply-chain-agents.html +166 -0
package/docs/blog/supply-chain-agents.md +79 -0
package/docs/business/index.html +530 -0
package/docs/business/install.html +247 -0
package/docs/checklist.html +168 -0
package/docs/finance/index.html +217 -0
package/docs/hall-of-fame.html +168 -0
package/docs/index.html +328 -90
package/docs/install.sh +557 -0
package/docs/privacy-policy/index.html +122 -0
package/docs/scan/index.html +214 -0
package/docs/sitemap.xml +132 -2
package/docs/support/index.html +124 -0
package/docs/terms-of-service/index.html +122 -0
package/examples/basic-usage.js +38 -0
package/package.json +1 -1
package/server/index.js +179 -14
package/server/index.js.patch +1 -0
package/src/finance/index.js +585 -0
package/src/finance/mcp-firewall.js +486 -0
package/src/guardian/cve-verify.js +129 -0
package/src/guardian/gateway-monitor.js +590 -0
package/src/guardian/index.js +3 -1
package/src/guardian/insider-threat.js +498 -0
package/src/index.js +3 -0
package/src/middleware/openclaw.js +28 -1

package/docs/blog/ollama-openclaw-security.html ADDED Viewed

@@ -0,0 +1,154 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+<meta charset="UTF-8">
+<meta name="viewport" content="width=device-width, initial-scale=1.0">
+<title>Ollama Just Made OpenClaw One-Click. Here's How to Secure It. | ClawMoat</title>
+<meta name="description" content="Ollama 0.17 ships native OpenClaw integration. Great for adoption — terrifying for security. Here's what you need to know.">
+<meta property="og:title" content="Ollama Just Made OpenClaw One-Click. Here's How to Secure It.">
+<meta property="og:description" content="Ollama 0.17 ships native OpenClaw integration with web search. More installs = more exposed hosts. Here's how to lock it down.">
+<link rel="canonical" href="https://clawmoat.com/blog/ollama-openclaw-security.html">
+<style>
+:root{--bg:#0a0a0f;--fg:#e0e0e8;--accent:#00d4aa;--muted:#888;--card:#14141f;--red:#ff4444}
+*{margin:0;padding:0;box-sizing:border-box}
+body{background:var(--bg);color:var(--fg);font-family:-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,sans-serif;line-height:1.8}
+.container{max-width:750px;margin:0 auto;padding:2rem 1.5rem}
+nav{padding:1rem 0;border-bottom:1px solid #2a2a3a;margin-bottom:2rem}
+nav a{color:var(--fg);text-decoration:none;margin-right:1.5rem}
+nav a:hover{color:var(--accent)}
+h1{font-size:2.2rem;line-height:1.2;margin-bottom:1rem}
+h2{color:var(--accent);margin:2rem 0 1rem;font-size:1.5rem}
+h3{margin:1.5rem 0 .75rem}
+p{margin-bottom:1rem}
+a{color:var(--accent)}
+pre{background:#1a1a2e;padding:1.25rem;border-radius:8px;overflow-x:auto;margin:1rem 0}
+code{background:#1a1a2e;padding:.15em .4em;border-radius:4px;font-size:.9em}
+pre code{background:none;padding:0}
+.meta{color:var(--muted);margin-bottom:2rem}
+.alert{background:#2a1a1a;border-left:4px solid var(--red);padding:1rem 1.25rem;margin:1.5rem 0;border-radius:0 8px 8px 0}
+.tip{background:#1a2a1a;border-left:4px solid var(--accent);padding:1rem 1.25rem;margin:1.5rem 0;border-radius:0 8px 8px 0}
+ul,ol{margin:1rem 0 1rem 1.5rem}
+li{margin-bottom:.5rem}
+blockquote{border-left:3px solid var(--muted);padding-left:1rem;color:var(--muted);margin:1rem 0}
+.cta{background:var(--accent);color:#000;padding:.75rem 1.5rem;border-radius:6px;text-decoration:none;font-weight:700;display:inline-block;margin:1rem .5rem 1rem 0}
+</style>
+</head>
+<body>
+<div class="container">
+<nav>
+<a href="/" style="font-weight:bold">ClawMoat</a>
+<a href="/blog/">Blog</a>
+<a href="/finance/">Finance</a>
+<a href="https://github.com/darfaz/clawmoat">GitHub</a>
+</nav>
+<h1>Ollama Just Made OpenClaw One-Click. Here's How to Secure It.</h1>
+<p class="meta">February 27, 2026 · 5 min read</p>
+<p>Ollama 0.17 just shipped <strong>native OpenClaw integration</strong> with web search out of the box. Two commands and you have a personal AI agent running on your machine with local models.</p>
+<p>This is great for adoption. It's terrifying for security.</p>
+<h2>What Ollama 0.17 Does</h2>
+<p>Ollama's latest release lets you set up OpenClaw to work with open models (Llama, Mistral, DeepSeek, etc.) and web search. No cloud API keys needed. Fully local inference.</p>
+<pre><code>ollama launch openclaw</code></pre>
+<p>That's it. One command. You now have an AI agent that can:</p>
+<ul>
+<li>Send emails on your behalf</li>
+<li>Manage your calendar</li>
+<li>Read and write files on your machine</li>
+<li>Execute shell commands</li>
+<li>Search the web</li>
+<li>Connect to WhatsApp, Telegram, iMessage</li>
+</ul>
+<p>All running with <strong>your user permissions</strong>. On <strong>your actual machine</strong>.</p>
+<h2>Why This Is a Security Problem</h2>
+<div class="alert">
+<strong>⚠️ The Ollama + OpenClaw combo inherits every OpenClaw vulnerability.</strong> Local models don't fix host-level security.
+</div>
+<p>Running local models solves one problem (data doesn't leave your machine) but creates a false sense of security. Here's what's still exposed:</p>
+<h3>1. Your Entire Filesystem</h3>
+<p>The agent runs as your user. It can read <code>~/.ssh</code>, <code>~/.aws</code>, browser cookies, crypto wallets, tax documents — everything you can access.</p>
+<h3>2. The WebSocket Hijack (CVE-2026-25253)</h3>
+<p>OpenClaw's gateway listens on localhost WebSocket. <a href="/blog/oasis-websocket-hijack.html">Oasis Security proved</a> any website can brute-force the port and take full control of your agent. Local models don't change this — the gateway architecture is the same.</p>
+<h3>3. Prompt Injection via Web Search</h3>
+<p>Ollama 0.17 adds web search. That means the agent fetches content from the internet and processes it. A malicious webpage can embed prompt injection payloads that hijack the agent's behavior. Now your "local" agent is executing attacker instructions.</p>
+<h3>4. Skill Supply Chain</h3>
+<p>OpenClaw skills are npm packages or GitHub repos. <a href="/blog/40000-exposed-openclaw-instances.html">341+ malicious skills</a> have been documented. A compromised skill runs with full access to your system.</p>
+<h3>5. No Permission Boundaries</h3>
+<p>OpenClaw has no concept of "this agent can read files but not execute commands" or "this agent can access the calendar but not SSH keys." It's all-or-nothing.</p>
+<h2>The One-Click Problem</h2>
+<p>When something is easy to install, people don't think about security. Ollama's user base is developers and tinkerers who want to run AI locally — they're not enterprise security teams. They'll run <code>ollama launch openclaw</code>, connect it to WhatsApp, and forget about it.</p>
+<blockquote>Microsoft: "OpenClaw should be treated as untrusted code execution with persistent credentials. It is not appropriate to run on a standard personal or enterprise workstation."</blockquote>
+<p>Now Ollama is making it trivial to do exactly what Microsoft says not to do.</p>
+<h2>How to Secure Your Ollama + OpenClaw Setup</h2>
+<div class="tip">
+<strong>✅ ClawMoat adds the security layer that Ollama + OpenClaw are missing.</strong>
+</div>
+<pre><code>npm install -g clawmoat</code></pre>
+<h3>1. Set Up Permission Tiers</h3>
+<pre><code>const { HostGuardian } = require('clawmoat');
+const guardian = new HostGuardian({
+  mode: 'standard',  // observer → worker → standard → full
+  workspace: '~/openclaw-workspace',
+  forbiddenZones: ['~/.ssh', '~/.aws', '~/.gnupg'],
+});</code></pre>
+<h3>2. Monitor Network Egress</h3>
+<pre><code>const { NetworkEgressLogger } = require('clawmoat');
+const logger = new NetworkEgressLogger();
+// Blocks requests to cloud metadata, private IPs, known-bad domains
+// Alerts on unusual outbound connections</code></pre>
+<h3>3. Scan Skills Before Installing</h3>
+<pre><code># Audit all installed skills for suspicious patterns
+npx clawmoat skill-audit ~/.openclaw/skills/</code></pre>
+<h3>4. Detect WebSocket Hijack Attempts</h3>
+<pre><code>const { GatewayMonitor } = require('clawmoat');
+const monitor = new GatewayMonitor();
+// Detects brute-force port scanning, suspicious WS origins,
+// unauthorized device pairing attempts</code></pre>
+<h3>5. Protect Financial Data</h3>
+<pre><code>const { FinanceGuard } = require('clawmoat');
+const guard = new FinanceGuard();
+// Blocks access to crypto wallets, banking files, tax documents
+// Redacts financial secrets in agent output</code></pre>
+<h2>The Bottom Line</h2>
+<p>Ollama 0.17 is going to put OpenClaw on thousands of new machines. Most of those machines won't have any security layer between the agent and the host.</p>
+<p><strong>If you're going to run OpenClaw — with Ollama or otherwise — run it with a security moat.</strong></p>
+<p>
+<a href="https://github.com/darfaz/clawmoat" class="cta">⭐ Star on GitHub</a>
+<a href="/#pricing" class="cta" style="background:#1a1a2e;color:var(--accent);border:2px solid var(--accent)">Get Started Free</a>
+</p>
+<p style="color:var(--muted);margin-top:2rem;font-size:.9rem">ClawMoat is open-source (MIT), zero dependencies, 277 tests passing. Works with any OpenClaw deployment — cloud, local, or Ollama.</p>
+</div>
+</body>
+</html>

package/docs/blog/openclaw-enterprise-readiness-claw10.html ADDED Viewed

@@ -0,0 +1,198 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+<meta charset="UTF-8">
+<meta name="viewport" content="width=device-width, initial-scale=1.0">
+<title>OpenClaw Scores 1.2/5 for Enterprise Readiness. Here's How to Fix 4 of the 10 Gaps. | ClawMoat Blog</title>
+<meta name="description" content="Onyx AI's CLAW-10 framework scored OpenClaw 1.2 out of 5 for enterprise readiness. ClawMoat directly addresses 4 of the 10 gaps: authorization, audit logging, privilege model, and supply chain security.">
+<meta name="keywords" content="OpenClaw enterprise readiness, CLAW-10 framework, AI agent enterprise security, OpenClaw compliance, ClawMoat enterprise, OpenClaw authorization, OpenClaw audit trail">
+<link rel="canonical" href="https://clawmoat.com/blog/openclaw-enterprise-readiness-claw10.html">
+<meta property="og:title" content="OpenClaw Scores 1.2/5 for Enterprise Readiness. ClawMoat Fixes 4 of the 10 Gaps.">
+<meta property="og:description" content="Onyx AI's CLAW-10 scored OpenClaw at 1.2/5. ClawMoat addresses authorization, audit logging, privilege model, and supply chain. Here's the mapping.">
+<meta property="og:url" content="https://clawmoat.com/blog/openclaw-enterprise-readiness-claw10.html">
+<meta property="og:type" content="article">
+<link rel="icon" href="data:image/svg+xml,<svg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 100 100'><text y='.9em' font-size='90'>🏰</text></svg>">
+<style>
+*{margin:0;padding:0;box-sizing:border-box}
+:root{--navy:#0F172A;--navy-light:#1E293B;--navy-mid:#334155;--blue:#3B82F6;--emerald:#10B981;--white:#F8FAFC;--gray:#94A3B8;--red:#EF4444;--amber:#F59E0B}
+body{font-family:-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,sans-serif;background:var(--navy);color:var(--white);line-height:1.8}
+a{color:var(--blue)}
+.container{max-width:740px;margin:0 auto;padding:0 24px}
+nav{background:rgba(15,23,42,.95);backdrop-filter:blur(12px);border-bottom:1px solid rgba(59,130,246,.15);padding:16px 0;position:fixed;top:0;left:0;right:0;z-index:100}
+nav .container{display:flex;align-items:center;justify-content:space-between}
+.logo{font-size:1.1rem;font-weight:700;color:var(--white);text-decoration:none}
+.logo span{color:var(--emerald)}
+nav a{color:var(--gray);font-size:.85rem;text-decoration:none}
+nav a:hover{color:var(--white)}
+article{padding:120px 0 80px}
+.meta{color:var(--gray);font-size:.85rem;margin-bottom:32px}
+h1{font-size:clamp(1.8rem,4vw,2.4rem);font-weight:800;line-height:1.2;margin-bottom:16px}
+h2{font-size:1.3rem;font-weight:700;margin:40px 0 16px}
+h3{font-size:1.05rem;font-weight:600;margin:28px 0 12px}
+p{color:var(--gray);margin-bottom:20px;font-size:1rem}
+table{width:100%;border-collapse:collapse;margin:24px 0;font-size:.9rem}
+th{text-align:left;padding:12px;color:var(--gray);border-bottom:1px solid var(--navy-mid);font-size:.8rem;text-transform:uppercase;letter-spacing:.05em}
+td{padding:12px;border-bottom:1px solid rgba(255,255,255,.05)}
+.score{display:inline-block;padding:2px 10px;border-radius:12px;font-weight:700;font-size:.85rem}
+.score-red{background:rgba(239,68,68,.15);color:var(--red)}
+.score-amber{background:rgba(245,158,11,.15);color:var(--amber)}
+.score-green{background:rgba(16,185,129,.15);color:var(--emerald)}
+blockquote{border-left:3px solid var(--blue);padding:16px 24px;margin:24px 0;background:var(--navy-light);border-radius:0 8px 8px 0}
+blockquote p{color:var(--white);margin:0;font-style:italic}
+blockquote cite{display:block;color:var(--gray);font-size:.85rem;margin-top:8px;font-style:normal}
+code{background:var(--navy-light);padding:2px 6px;border-radius:4px;font-size:.9rem;color:var(--emerald)}
+pre{background:#0a0e17;border:1px solid var(--navy-mid);border-radius:8px;padding:20px;overflow-x:auto;margin:24px 0;font-size:.85rem}
+pre code{background:none;padding:0}
+ul,ol{color:var(--gray);margin:0 0 20px 24px}
+li{margin-bottom:8px}
+.cta{background:linear-gradient(135deg,rgba(16,185,129,.1),rgba(59,130,246,.1));border:1px solid rgba(16,185,129,.2);border-radius:12px;padding:32px;text-align:center;margin:48px 0}
+.cta h3{margin:0 0 12px;color:var(--white)}
+.cta p{margin:0 0 20px}
+.cta a{display:inline-block;background:var(--emerald);color:#fff;padding:12px 28px;border-radius:8px;font-weight:600;text-decoration:none}
+</style>
+</head>
+<body>
+<nav>
+<div class="container">
+  <a href="/" class="logo">🏰 Claw<span>Moat</span></a>
+  <div style="display:flex;gap:20px">
+    <a href="/blog/">Blog</a>
+    <a href="/#features">Features</a>
+    <a href="/business/">For Business</a>
+    <a href="https://github.com/darfaz/clawmoat">GitHub</a>
+  </div>
+</div>
+</nav>
+<article>
+<div class="container">
+<div class="meta">February 26, 2026 · 6 min read · By the ClawMoat Team</div>
+<h1>OpenClaw Scores 1.2/5 for Enterprise Readiness. Here's How to Fix 4 of the 10 Gaps.</h1>
+<p style="font-size:1.1rem;color:var(--white)">Onyx AI just published the <a href="https://www.onyx.app/insights/openclaw-enterprise-evaluation-framework">CLAW-10 Enterprise Readiness Matrix</a> — the first structured framework for evaluating autonomous AI agents against enterprise requirements. OpenClaw scored <strong>1.2 out of 5</strong>. The enterprise-ready threshold is 4.0.</p>
+<p>Every single dimension scored below the threshold. But here's the thing: 4 of those 10 gaps can be addressed today with a single npm install.</p>
+<h2>The CLAW-10 Scores</h2>
+<table>
+<thead>
+<tr><th>#</th><th>Dimension</th><th>OpenClaw Score</th><th>With ClawMoat</th><th>Threshold</th></tr>
+</thead>
+<tbody>
+<tr><td>1</td><td>Identity & Authentication</td><td><span class="score score-red">1/5</span></td><td><span class="score score-red">1/5</span></td><td>4</td></tr>
+<tr><td>2</td><td><strong>Authorization & Access Control</strong></td><td><span class="score score-red">1/5</span></td><td><span class="score score-amber">3/5</span></td><td>4.5</td></tr>
+<tr><td>3</td><td><strong>Audit Logging & Observability</strong></td><td><span class="score score-amber">2/5</span></td><td><span class="score score-green">4/5</span></td><td>4.5</td></tr>
+<tr><td>4</td><td>Data Isolation & Residency</td><td><span class="score score-red">1/5</span></td><td><span class="score score-red">1.5/5</span></td><td>4</td></tr>
+<tr><td>5</td><td>Execution Sandboxing</td><td><span class="score score-red">1/5</span></td><td><span class="score score-amber">2/5</span></td><td>4.5</td></tr>
+<tr><td>6</td><td>Compliance Certifications</td><td><span class="score score-red">1/5</span></td><td><span class="score score-red">1.5/5</span></td><td>4</td></tr>
+<tr><td>7</td><td><strong>Supply Chain Security</strong></td><td><span class="score score-red">1/5</span></td><td><span class="score score-amber">2.5/5</span></td><td>4</td></tr>
+<tr><td>8</td><td>Network Exposure & Attack Surface</td><td><span class="score score-amber">2/5</span></td><td><span class="score score-amber">3.5/5</span></td><td>4</td></tr>
+<tr><td>9</td><td><strong>Privilege Model</strong></td><td><span class="score score-red">1/5</span></td><td><span class="score score-amber">3.5/5</span></td><td>4</td></tr>
+<tr><td>10</td><td>Vendor Support & SLAs</td><td><span class="score score-red">1/5</span></td><td><span class="score score-red">1/5</span></td><td>3</td></tr>
+<tr style="font-weight:700;border-top:2px solid var(--navy-mid)"><td></td><td>Composite</td><td><span class="score score-red">1.2/5</span></td><td><span class="score score-amber">2.2/5</span></td><td>4.0</td></tr>
+</tbody>
+</table>
+<p>ClawMoat raises the composite score from 1.2 to 2.2 — not enterprise-ready yet, but a significant improvement. Let's look at the four dimensions where ClawMoat makes the biggest difference.</p>
+<h2>1. Authorization & Access Control (1 → 3)</h2>
+<p>OpenClaw's current score: <strong>1/5</strong>. There is no authorization layer. The agent runs with the user's full permissions.</p>
+<p>With ClawMoat's Host Guardian, you get four permission tiers:</p>
+<ul>
+<li><strong>Observer</strong> — read-only. Can list files and view system info. Cannot write or execute.</li>
+<li><strong>Worker</strong> — safe commands (git, npm, file I/O). Blocks dangerous operations.</li>
+<li><strong>Standard</strong> — most operations permitted. Forbidden zones enforced.</li>
+<li><strong>Full</strong> — unrestricted execution. Forbidden zones still active. Full audit trail.</li>
+</ul>
+<p>Plus 20+ forbidden zones that block access to sensitive directories regardless of tier: <code>~/.ssh</code>, <code>~/.aws</code>, <code>~/.gnupg</code>, browser data, crypto wallets, package tokens, and more.</p>
+<h2>2. Audit Logging & Observability (2 → 4)</h2>
+<p>OpenClaw's current score: <strong>2/5</strong>. Basic session logging exists, but there's no structured audit trail for security events.</p>
+<p>ClawMoat provides:</p>
+<ul>
+<li>Full audit trail of every file access, shell command, and network request</li>
+<li>Credential file monitoring (watches sensitive directories for unauthorized access)</li>
+<li>Network egress logging with URL extraction and domain tracking</li>
+<li>Real-time alerts via console, file, webhook, Slack, or email</li>
+<li>Exportable logs for compliance review and incident forensics</li>
+<li>Rate-limited alert delivery to prevent alert fatigue</li>
+</ul>
+<p>This is the dimension where ClawMoat makes the biggest impact — taking OpenClaw from "basic logs" to "structured, exportable, real-time observability."</p>
+<h2>3. Privilege Model (1 → 3.5)</h2>
+<p>OpenClaw's current score: <strong>1/5</strong>. No privilege model exists. The agent inherits the user's full privileges.</p>
+<p>ClawMoat's permission tiers implement least privilege at the host level:</p>
+<ul>
+<li>Tiered command blocking — dangerous commands (rm -rf, chmod 777, etc.) blocked by tier</li>
+<li>File system restrictions — read/write access controlled per tier</li>
+<li>Forbidden zones — always blocked, regardless of tier</li>
+<li>Runtime tier switching — promote or demote without restart</li>
+</ul>
+<h2>4. Supply Chain Security (1 → 2.5)</h2>
+<p>OpenClaw's current score: <strong>1/5</strong>. Skills are installed from ClawHub with no signature verification.</p>
+<p>ClawMoat's skill integrity checker provides:</p>
+<ul>
+<li>Hash-based verification of installed skills</li>
+<li>14 suspicious pattern detectors (eval, exec, fetch to unknown domains, etc.)</li>
+<li>CLI: <code>clawmoat skill-audit ~/.openclaw/skills/</code></li>
+<li>Detects tampering after installation</li>
+</ul>
+<p>This doesn't solve the root problem (no signature verification in ClawHub), but it provides a detection layer that didn't exist before.</p>
+<h2>What ClawMoat Doesn't Fix</h2>
+<p>Honesty matters. ClawMoat doesn't address:</p>
+<ul>
+<li><strong>Identity & Authentication (1/5)</strong> — OpenClaw has no agent identity system. This needs to be fixed upstream.</li>
+<li><strong>Data Isolation (1/5)</strong> — The agent still runs in the user's environment. True data isolation requires VM-level separation.</li>
+<li><strong>Execution Sandboxing (1/5)</strong> — ClawMoat restricts what the agent can do, but doesn't sandbox execution at the process level. (See <a href="https://news.ycombinator.com/item?id=47075823">ClawShell</a> for process-level isolation.)</li>
+<li><strong>Compliance Certifications (1/5)</strong> — ClawMoat generates compliance-ready reports, but isn't itself certified.</li>
+<li><strong>Vendor Support (1/5)</strong> — We're open source. Enterprise support is available through our <a href="/business/">business plans</a>.</li>
+</ul>
+<h2>The Defense-in-Depth Stack</h2>
+<p>No single tool gets OpenClaw to 4.0/5. The enterprise stack will look something like:</p>
+<table>
+<thead><tr><th>Layer</th><th>Tool</th><th>CLAW-10 Dimensions</th></tr></thead>
+<tbody>
+<tr><td>Host Security</td><td><strong>ClawMoat</strong></td><td>Authorization, Audit, Privilege, Supply Chain</td></tr>
+<tr><td>Process Isolation</td><td>ClawShell</td><td>Data Isolation, Execution Sandboxing</td></tr>
+<tr><td>Prompt Scanning</td><td>LlamaFirewall</td><td>Network Exposure (input filtering)</td></tr>
+<tr><td>Enterprise Governance</td><td>Runlayer</td><td>Identity, Compliance, Vendor Support</td></tr>
+<tr><td>Managed Hosting</td><td>KiloClaw</td><td>Execution Sandboxing, Network Exposure</td></tr>
+</tbody>
+</table>
+<div class="cta">
+<h3>Start improving your CLAW-10 score today.</h3>
+<p>One npm install. Four dimensions addressed. Zero dependencies.</p>
+<a href="https://github.com/darfaz/clawmoat">⭐ Star on GitHub</a>
+</div>
+<pre><code>npm install -g clawmoat
+clawmoat scan ~/.openclaw/
+clawmoat skill-audit ~/.openclaw/skills/
+clawmoat report</code></pre>
+<p>For businesses that want managed installation + compliance reports: <a href="/business/">ClawMoat for Business →</a></p>
+<p style="color:var(--gray);font-size:.85rem;margin-top:48px;padding-top:24px;border-top:1px solid rgba(255,255,255,.06)">Sources: <a href="https://www.onyx.app/insights/openclaw-enterprise-evaluation-framework">Onyx AI CLAW-10</a> · <a href="https://www.microsoft.com/en-us/security/blog/2026/02/19/running-openclaw-safely-identity-isolation-runtime-risk/">Microsoft Security Blog</a> · <a href="https://www.catonetworks.com/blog/when-ai-can-act-governing-openclaw/">Cato Networks</a></p>
+</div>
+</article>
+</body>
+</html>