> For the complete documentation index, see [llms.txt](https://docs.vex-nx.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.vex-nx.com/developers/mcp-and-ai-tools.md).

# MCP and AI tools

![Assist / MCP tooling surface](https://1807546539-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FhgHCHWL3861o2DBW53aA%2Fuploads%2FeuxaIAw2pvhAX5CWNo4e%2Fve20261112v5-assist-studio.png?alt=media)

## Goal

Understand how VEX-NX exposes lab tools to Assist and external AI agents today.

## What ships today

VEX-NX publishes an **MCP-style tool catalog over REST** (not a desktop stdio / SSE MCP server process):

| Method | Path                | Purpose                                      |
| ------ | ------------------- | -------------------------------------------- |
| GET    | `/api/v1/mcp/tools` | List tool names, descriptions, input schemas |
| POST   | `/api/v1/mcp/call`  | Execute one allowlisted tool                 |

Available on both editions (Student and Enterprise - see [Features](/getting-started/features.md)); requires a valid JWT.

Assist uses the same allowlisted tools server-side when planning lab actions.

## What this is not (yet)

* Not Cisco CML's packaged MCP desktop integration story
* Not a local `npx` / stdio MCP binary you attach in Cursor settings without HTTP
* Agents should call the appliance HTTPS/HTTP API with a user token

## Tool catalog (names)

Examples of tools returned by `GET /api/v1/mcp/tools` (names may grow with releases):

| Tool                                                                               | Role                          |
| ---------------------------------------------------------------------------------- | ----------------------------- |
| `list_labs` / `get_lab` / `create_lab`                                             | Lab inventory                 |
| `add_node` / `patch_node` / `add_link` / `delete_link`                             | Topology edits                |
| `validate_lab` / `assist_lab`                                                      | Validate and Q\&A             |
| `list_templates` / `list_presets`                                                  | Catalog                       |
| `generate_lab_configs` / `apply_lab_configs`                                       | Config proposals              |
| `start_lab_nodes`                                                                  | Power                         |
| `preview_running_configs` / `create_lab_from_running_configs`                      | Map / Remap / Redesign import |
| `node_exec` / `node_get_running_config` / `node_apply_config` / `node_save_config` | Live CLI                      |
| `start_capture` / `list_captures`                                                  | Packet capture                |
| `assist_agent`                                                                     | Orchestrated Assist run       |
| `export_lab_configs`                                                               | Config pack path              |
| Preset guide / publish override tools                                              | Instructors / admins          |

Always fetch `/api/v1/mcp/tools` from your appliance for the live schema.

## Call shape

```http
POST /api/v1/mcp/call
Authorization: Bearer <token>
Content-Type: application/json

{
  "name": "list_labs",
  "arguments": {}
}
```

Never paste real Bearer tokens into tickets or public repos.

## Example: list tools

```bash
curl -fsS "http://<appliance-ip>/api/v1/mcp/tools" \
  -H "Authorization: Bearer <token>"
```

## Using with ChatGPT / Claude / Cursor

1. Point the agent at your appliance base URL (reachable from that environment)
2. Provide a short-lived user JWT (or a script that logs in)
3. Prefer `GET /api/v1/mcp/tools` then `POST /api/v1/mcp/call`
4. Or drive the UI Assist dock for interactive confirms

Example agent prompt:

> Using the VEX-NX API at http\://, list MCP tools, create an empty lab named demo-ospf, add three cisco-iosv nodes if templates exist, and do not start nodes until I confirm.

## Related

* REST API overview
* [Using Assist](/assist/using-assist.md)
* Recreate labs from production configs
* OpenAPI reference


---

# 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://docs.vex-nx.com/developers/mcp-and-ai-tools.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.