# Plan a mission with an AI assistant, on real data

Source: https://onvector.eu/ai-mission-planning - last updated 2026-10-22

Vector has an MCP server, so an AI assistant such as Claude or Codex can work on a mission plan. The assistant reads the same plan the team sees, with its orders, and answers from real data: weather for the planned hour, airspace, terrain, line of sight and routes. What it adds appears on the map for everyone in the mission.

## What the assistant can do

- Read the mission: objects, layers, members and orders, in one call for the whole situation.
- Check weather, airspace, terrain, line of sight, viewshed and march time.
- Plan a road route between two points and add it to the map.
- Check the whole plan against the orders.
- Create missions and add, change or delete objects, within the scopes of its key.

## An example

Asked "what are the conditions at H-hour?" for a demo mission near Ādaži, Latvia, the assistant called the weather and airspace tools and answered: GO at 06:00, 7.6 °C, wind 3.7 m/s with gusts of 7.3, the window closes at 09:00 when cloud reaches 96%; the launch site is in the Ādaži ATZ, so notify the aviation authority 2 hours ahead.

## How to connect

Create an API key in Settings with only the scopes it needs. Add Vector to your MCP client with the address https://onvector.eu/api/mcp and the key as a Bearer token. The same key works for the REST API. Keys can be revoked at any time, and every request made with a key is logged.

## Price

API and MCP keys are part of Pro: EUR 6 per user a month, or EUR 60 a year.

## Questions

### Does Vector have an AI chat inside the app?

No. Vector connects to the AI assistant you already use, through MCP.

### Which AI assistants work with Vector?

Claude, Codex and any other client that supports MCP over Streamable HTTP.