mcp MCP
v0.1.0dev.safeprompt/mcp
Detect prompt injection, jailbreaks, and code injection in untrusted text before it reaches an LLM.
transport:stdio
runtime:npm
Target client
run in your project directory
claude mcp add mcp -e SAFEPROMPT_API_KEY=<SAFEPROMPT_API_KEY> -e SAFEPROMPT_PROVIDER=<SAFEPROMPT_PROVIDER> -e SAFEPROMPT_USER_IP=<SAFEPROMPT_USER_IP> -- npx -y @safeprompt.dev/mcp
Adds it for this project only. Append --scope user to make it available everywhere. Claude Code docs
This listing does not declare its tools. Connect the server and your client will discover them on the handshake.