# tool-use-firewall MCP server

> Acts as a security proxy between AI agents and MCP servers to enforce policies like rate limiting, argument validation, and human-in-the-loop approvals.

- Repository: https://github.com/reaatech/tool-use-firewall
- Page: https://hostedmcp.org/servers/reaatech/tool-use-firewall
- Transport: —
- Stars: 1
- License: MIT

## Environment variables

| Name | Required | Secret | Description |
|---|---|---|---|
| `APPROVAL_API_TOKEN` | no | no |  |

## FAQ

### What does tool-use-firewall do?

It serves as a policy enforcement layer between AI agents and MCP servers, intercepting and validating tool calls for security and cost control.

### How can I approve dangerous tool calls?

The system supports human-in-the-loop approval workflows through multiple interfaces including HTTP API, CLI prompts, webhooks, Slack, and Discord.
