AgentDiet A2A Indirect Prompt-Injection & Jailbreak Firewall for Claude.ai

dev.agentdiet/prompt-guard

Scans untrusted web pages, emails, MCP responses, or A2A payloads for hidden prompt injections an.

client:Claude.ai transport:streamable-http tools:2

Install AgentDiet A2A Indirect Prompt-Injection & Jailbreak Firewall in Claude.ai

Settings → Connectors → Add custom connector

https://prompt-guard.agentdiet.dev/mcp

Claude on the web takes a remote URL and completes OAuth in the browser. For a packaged server use Claude Desktop or Claude Code instead. Claude.ai docs

What Claude.ai can do once it is connected

scan_prompt_injection
get_agent_wallet_status

AgentDiet A2A Indirect Prompt-Injection & Jailbreak Firewall in other clients

Claude Code
Claude Desktop
Cursor
VS Code
Zed
Windsurf
Cline
Gemini CLI
Grok
ChatGPT
LangChain

AgentDiet A2A Indirect Prompt-Injection & Jailbreak Firewall MCP server