# ClawGuard Shield

AI Agent Security Scanner — detect prompt injection, jailbreaks, and tool poisoning in real-time. 204+ attack patterns, 15 languages, EU AI Act compliance-ready. Scan text or MCP server configs for t…

## Quick Start

```bash
# Connect this server (installs CLI if needed)
npx -y @smithery/cli@latest mcp add joergmichno/clawguard

# Browse available tools
npx -y @smithery/cli@latest tool list joergmichno/clawguard

# Get full schema for a tool
npx -y @smithery/cli@latest tool get joergmichno/clawguard scan_text

# Call a tool
npx -y @smithery/cli@latest tool call joergmichno/clawguard scan_text '{}'
```

## Direct MCP Connection

Endpoint: `https://clawguard--joergmichno.run.tools`

## Tools (2)

- `scan_text` — Scan text for prompt injection attacks, jailbreak attempts, and other AI security threats. Detects 204+ attack patterns…
- `scan_mcp_config` — Scan an MCP server configuration for security vulnerabilities: tool poisoning, permission escalation, data exfiltration…

```bash
# Get full input/output schema for a tool
npx -y @smithery/cli@latest tool get joergmichno/clawguard <tool-name>
```