CougarMap MCP server

CougarMap

CougarMap finds where to put a trail camera for mountain lions (cougars). Give it a place and it reads the terrain from free public map data. It ranks camera spots you can walk to and explains why each one is good. The results come as a Google Earth map.

It works anywhere in the United States. It was built and tuned in the dry conifer mountains of the inland Northwest.

Quick start

1. Install. On a Mac or Linux, in a terminal:

curl -LsSf https://raw.githubusercontent.com/bryantclark/cougarmap/main/installer/install.sh | sh

On Windows, in PowerShell:

powershell -c "irm https://raw.githubusercontent.com/bryantclark/cougarmap/main/installer/install.ps1 | iex"

It installs uv (which brings its own Python) if you don't have it, installs CougarMap, and connects it to the AI apps it finds (step 3). Run it again any time to update. Then open a new terminal window so the cougarmap command is found.

latest release and double-click the installer in it.

• Already have uv? uv tool install --python 3.12 cougarmap installs just the command.
• No terminal? Download CougarMap-installer.zip from the

2. Find camera spots around a coordinate. Give latitude and longitude the way Google Maps copies them. Optionally use the --radius-km flag to set a radius size.

cougarmap analyze --near "47.3712, -116.1029" --radius-km 3

The first run in a new place takes a few minutes, mostly downloading elevation and map data. Later runs take under a minute. When it finishes, it prints the best spots and a link to the map, and opens the map in Google Earth Pro (free).

To search a whole region around a town instead, use hotspots. It screens a 25 km circle and analyzes the most promising blocks:

cougarmap hotspots "Missoula, MT"

3. Use it from your AI assistant. The installer already connected CougarMap to the AI apps it found on your computer: Claude, Codex, Gemini CLI, Antigravity or Cursor. Restart the app, then ask "find me cougar camera spots near Missoula, MT". If you installed with uv tool install, or added an app later, run:

cougarmap setup

What you get

For each spot:

valley meets the prevailing wind";

• a score from 0 to 100 (60 or more is strong);
• the reasons in plain English, for example "downwind end of a meadow edge, where cold air draining down the

• the walk from the nearest road that is open that month, with distance and time;
• the land it sits on (national forest, state land and so on), and its coordinates.

The Google Earth map has the ranked pins (click one for its reasons), walking routes, dawn and dusk air-flow arrows, saddles, and a layer for each factor you can switch on and off.

The ranked spots are on public land within 1 mile of walking from a road open that month. Private land is scored the same way and its best spots are found too, but kept apart: they're listed separately (P1, P2...) and sit in map layers that start switched off. Tick "Private land spots" in Google Earth to see them, for example to scan your own property, and get landowner permission before using one. --max-walk-miles and --month change the other defaults.

How it picks spots

CougarMap stacks four things lions use. The more of them at one spot, the better the spot:

Shortened. The full README is on GitHub.