# mcp MCP server

Detect prompt injection, jailbreaks, and code injection in untrusted text before it reaches an LLM.

## Links
- Registry page: https://www.getdrio.com/mcp/dev-safeprompt-mcp
- Repository: https://github.com/ianreboot/safeprompt
- Website: https://safeprompt.dev

## Install
- Command: `npx -y @safeprompt.dev/mcp`
- Auth: Auth required by registry metadata

## Setup notes
- Package: Npm @safeprompt.dev/mcp v0.1.0
- Environment variable: SAFEPROMPT_API_KEY (required; secret)
- Environment variable: SAFEPROMPT_PROVIDER
- Environment variable: SAFEPROMPT_USER_IP
- The upstream registry signals required auth or secrets.

## Tools
Not captured

## Resources
Not captured

## Prompts
Not captured

## Metadata
- Owner: dev.safeprompt
- Version: 0.1.0
- Runtime: Npm
- Transports: STDIO
- License: Not captured
- Language: Not captured
- Stars: Not captured
- Updated: Jun 22, 2026
- Source: https://registry.modelcontextprotocol.io