MCP Setup

Audit AI agent skills for malware and social engineering before adding them to your agent.

What Shieldon Detects

Credential theft: AWS, GitHub, Stripe, Slack, cloud provider keys, private keys, and .env access
Command injection: curl | bash, reverse shells, privilege escalation
Obfuscation: Base64 payloads, eval() chains, hex encoding
Data exfiltration: POST to webhooks, paste services, DNS tunneling

Installation

Add Shieldon to your MCP config. Shieldon exposes an MCP Streamable HTTP endpoint — compatible with any MCP client that supports remote servers.

{
  "mcpServers": {
    "shieldon": {
      "url": "https://api.shieldon.dev/mcp"
    }
  }
}

Client	Config file
Cursor	.cursor/mcp.json
Claude Desktop	claude_desktop_config.json
Windsurf	.windsurf/mcp.json
Any MCP client	Wherever your client reads MCP server configs

After adding the config, restart your client or reload the session. The audit_skill tool will appear in your available MCP tools.

Usage

Scan a skill by URL before installing:

audit_skill(skill_url: "https://example.com/skill.md")

Scan raw skill content directly:

audit_skill(skill_content: "<paste skill.md content here>")

Example response:

{
  "scan_id": "a1b2c3d4",
  "risk_score": 75,
  "risk_level": "HIGH",
  "recommendation": "REVIEW",
  "findings": [
    {
      "type": "Credential_AWS_Access_Key",
      "description": "AWS access key pattern detected",
      "severity": "HIGH",
      "evidence": "Matched '$aws_key' at offset 342: AKIAIOSFODNN7EXAMPLE"
    }
  ],
  "scans_remaining": 7
}

Rate Limits

Free tier: 10 scans per day per IP
Privacy: Skills are never stored. All scans are ephemeral.

Support

Website: shieldon.dev