# humane-proxy

AI safety middleware — detects self-harm and criminal intent in LLM prompts.

- Weekly downloads: n/a
- GitHub stars: 28
- Health: active
- Repository: https://github.com/vishisht16/humane-proxy
- Registry names: io.github.Vishisht16/humane-proxy
- Categories: Dev tools & GitHub, AI models & media
- Snapshot: 2026-10-17

## Install

```sh
uvx humane-proxy
```

```json
{
  "mcpServers": {
    "humane-proxy": {
      "command": "uvx",
      "args": [
        "humane-proxy"
      ]
    }
  }
}
```

Source: https://mcpmorning.com/mcp/vishisht16-humane-proxy/