# EMA Prompt Injection Firewall v3.1.0

Real-time prompt injection and jailbreak detection for AI agents. Blocks instruction overrides, data exfiltration, tool poisoning and 8 attack types. Now with shared learning brain - confirmed attack…

## Quick Start

```bash
# Connect this server (installs CLI if needed)
npx -y @smithery/cli@latest mcp add ema-security/prompt-injection-firewall-v3-1-0

# Browse available tools
npx -y @smithery/cli@latest tool list ema-security/prompt-injection-firewall-v3-1-0

# Get full schema for a tool
npx -y @smithery/cli@latest tool get ema-security/prompt-injection-firewall-v3-1-0 scan

# Call a tool
npx -y @smithery/cli@latest tool call ema-security/prompt-injection-firewall-v3-1-0 scan '{}'
```

## Direct MCP Connection

Endpoint: `https://prompt-injection-firewall-v3-1-0--ema-security.run.tools`

## Tools (2)

- `scan` — Scan content for prompt injection, jailbreaks, tool poisoning, and encoding obfuscation. Returns CLEAN/SUSPICIOUS/BLOCK…
- `scan_batch` — Scan up to 20 items in one call at $0.02 USDC each.

```bash
# Get full input/output schema for a tool
npx -y @smithery/cli@latest tool get ema-security/prompt-injection-firewall-v3-1-0 <tool-name>
```