> For the complete documentation index, see [llms.txt](https://doc.threatreaper.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://doc.threatreaper.ai/user-guide/ai-mcp-mcp-security.md).

# 🔗 AI-MCP — MCP Security

**Menu:** PROTECT > AI-MCP **(limited availability — ask your administrator if this is enabled for your organisation)**.

AI-MCP extends LockThreat's protections to Model Context Protocol (MCP) servers and tool-calling agents, applying the same policy and guardrail model used by AI-Firewall to MCP requests and responses. If you don't see this module, it has not yet been enabled for your organisation or project.

## 9.1 How to configure it

**Settings (AI-MCP > Configuration):**

| **Field**                                       | **Purpose**                                                                  |
| ----------------------------------------------- | ---------------------------------------------------------------------------- |
| MCP Security Enabled                            | Master on/off switch for MCP protection                                      |
| Default Action                                  | What to do with a request that doesn't match any rule: Allow, Alert or Block |
| Log Level                                       | How much detail is recorded: Debug, Info, Warning or Error                   |
| Threat Detection Enabled                        | Turns on active threat scanning of MCP traffic                               |
| Rate Limiting Enabled + Max Requests Per Minute | Caps how many MCP requests are processed per minute                          |
| Allowed Tools / Blocked Tools                   | Explicit allow-list / block-list of MCP tool names                           |

**Registering a server (AI-MCP > Servers > Register):**

{% stepper %}
{% step %}

## Enter the server details

Enter a Server Name, Server URL and Version.
{% endstep %}

{% step %}

## Register the server

Select **Register**.
{% endstep %}
{% endstepper %}

## 9.2 Errors you might see and how to fix them

| **Message**                                 | **What it means**                                                                                        | **What to do**                                                               |
| ------------------------------------------- | -------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------- |
| A server shows status **Inactive**          | LockThreat hasn't received recent traffic from it, or it's unreachable                                   | Confirm the MCP server is running and the URL is correct                     |
| A tool call you expected to work is blocked | It's on the **Blocked Tools** list, or not on the **Allowed Tools** list while an allow-list is enforced | Review **AI-MCP > Configuration** and adjust the tool lists                  |
| Requests are being throttled                | Rate limiting is enabled and the per-minute cap was reached                                              | Raise **Max Requests Per Minute**, or confirm the traffic volume is expected |

## 9.3 What success looks like

**Servers** tab:

| **Server Name**    | **URL**               | **Status**  | **Version** | **Last Seen** |
| ------------------ | --------------------- | ----------- | ----------- | ------------- |
| mcp-internal-tools | mcp.internal.acme.com | 🟢 Active   | 1.4.2       | 2 minutes ago |
| mcp-db-connector   | mcp-db.acme.com       | 🔴 Inactive | 1.2.0       | 3 days ago    |

**Operations** tab shows total requests, allowed vs. blocked counts, and average response time, with a filterable log of individual tool calls and their decisions.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://doc.threatreaper.ai/user-guide/ai-mcp-mcp-security.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.