# PROMPT-GUARD — A2A Indirect Prompt-Injection & Jailbreak Firewall (mcp-security.agentdiet.dev)

> Scans untrusted web pages, emails, MCP responses, or A2A payloads for hidden prompt injections and returns sanitized text. (Live A2A v0.3, MCP & x402 v2 node at https://mcp-security.agentdiet.dev)

- **Domain**: `mcp-security.agentdiet.dev`
- **Provider**: AgentDiet B2A Infrastructure (https://mcp-security.agentdiet.dev)
- **A2A protocol version**: 1.0
- **Kind**: paid_api
- **Live-responds (last probe)**: True
- **Signed card**: True
- **Streaming**: False
- **Quality score**: 100%
- **Payment protocol**: x402 · network `eip155:8453`
- **Wallet**: `0xec5275d5b03bc390fca63d902d0b2d0db2eb5573`

## Skills
- **A2A Indirect Prompt-Injection & Jailbreak Firewall** — Scans untrusted web pages, emails, MCP responses, or A2A payloads for hidden prompt injections and returns sanitized text.
- **scan_prompt_injection (MCP JSON-RPC)** — Execute scan_prompt_injection directly over MCP Streamable HTTP JSON-RPC 2.0 at https://mcp-security.agentdiet.dev/mcp/prompt-guard.
- **Check Agent Wallet & Free-Trial Quota** — Inspect remaining free-trial calls, prepaid USDC session pass balance, and cumulative ROI savings.

## URLs
- Agent card: https://mcp-security.agentdiet.dev/.well-known/agent-card.json
- Page (HTML): https://agenstry.com/agents/mcp-security.agentdiet.dev
- Documentation: https://mcp-security.agentdiet.dev/llms.txt