Skip to content

CLI Reference

GeoMind is accessed through the geomind command. Once installed via pip install geomind-ai, the command is available globally in your terminal.


Basic Usage

# Launch interactive mode
geomind

# Run a single query
geomind --query "Find recent imagery of Paris"

# Use a specific model
geomind --model "anthropic/claude-3.5-sonnet"

Options

Flag Short Description Example
--help -h Show help message and exit geomind -h
--query QUERY -q Single query to run (if omitted, starts interactive mode) geomind -q "Find recent imagery of Paris"
--model MODEL -m Model name to use geomind -m "anthropic/claude-3.5-sonnet"
--api-key API_KEY -k OpenRouter API key (overrides saved key) geomind -k "sk-or-v1-..."
--version -v Show version and exit geomind -v
--clear-key - Clear the saved API key geomind --clear-key

Environment Variables

Variable Description Default
OPENROUTER_API_KEY Your OpenRouter API key. Get one free None
OPENROUTER_MODEL LLM model identifier nvidia/nemotron-3-nano-30b-a3b:free
OPENROUTER_API_URL OpenRouter API endpoint https://openrouter.ai/api/v1

Using a .env file

Create a .env file in your working directory:

OPENROUTER_API_KEY=sk-or-v1-your-key-here
OPENROUTER_MODEL=nvidia/nemotron-3-nano-30b-a3b:free
GeoMind loads this automatically on startup.


Operating Modes

Interactive Mode (Default)

When you run geomind without any arguments, it starts an interactive conversational session:

geomind
+----------------------------------------------------------------------+
| >_ GeoMind (v1.2.0)                                                  |
|                                                                      |
| model:     nvidia/nemotron-3-nano-30b-a3b:free                       |
| docs:      https://harshshinde0.github.io/GeoMind                    |
| authors:   Harsh Shinde, Rajat Shinde                                |
|                                                                      |
| Type "?" for help, "quit" to exit.                                   |
+----------------------------------------------------------------------+

GeoMind Agent initialized with nvidia/nemotron-3-nano-30b-a3b:free

>

You can type natural language queries continuously:

> Create an RGB composite for the most recent image of Iceland
  Executing: list_recent_imagery({'location_name': 'Iceland'})
  Executing: create_rgb_composite({...})

The RGB composite for the most recent Sentinel-2 image of Iceland has been
generated successfully.

- Output file: outputs/rgb_composite_254.png
- Image size: 1000 x 1000 px
- Bands used: Red (B04), Green (B03), Blue (B02)

> get me recent image of scotland and its ndvi
  Executing: list_recent_imagery({...})
  Executing: create_rgb_composite({...})
  Executing: calculate_ndvi({...})

RGB composite: outputs/rgb_composite_7066.png
NDVI:          outputs/ndvi_9664.png
NDVI stats:    min, max, mean, std

>

Multi-turn Conversation

The agent retains memory of the current session. You can ask follow-up questions like "Now calculate NDVI for that same scene" and it will reuse the previous context.

Single-Query Mode

For scripting or one-off queries, use the --query flag:

geomind --query "Find recent imagery of Paris"

This executes the query, prints results, and exits.

geomind -q "Create an RGB true-color composite visualization from the December 25th image"
geomind -q "Create an NDVI from the December 25th image"

API Key Management

First-Time Setup

On the very first launch, GeoMind will ask for your API key:

OpenRouter API key required (FREE)
   Get yours at: https://openrouter.ai/settings/keys

   Enter your API key: sk-or-v1-xxxxxxxxxxxxxxxxxxxxxxxx
   API key saved! You won't need to enter it again.

The key is saved to ~/.geomind/config and is loaded automatically on future runs.

Key Priority

GeoMind checks for the API key in this order:

  1. --api-key command-line flag (highest priority)
  2. OPENROUTER_API_KEY environment variable
  3. Saved key in ~/.geomind/config

Clearing the Key

geomind --clear-key
Saved API key cleared.


Interactive Session Commands

Once inside an interactive session, these special commands are available:

Command Action
help or ? Display the built-in help text
reset Wipe conversation memory and reset context
exit, quit, or q Close GeoMind

Built-in Help Text

> ?

Interactive Commands:
  help, ?          Show this help
  reset            Reset conversation
  exit, quit, q    Exit GeoMind

Query Examples:
  > Find recent Sentinel-2 imagery of Paris
  > Show me NDVI data for the Amazon rainforest
  > Search for images with less than 10% cloud cover in London
  > Get satellite data for coordinates 40.7128, -74.0060

For CLI options, run: geomind --help

Processing Indicator

When processing a query, GeoMind shows a real-time spinner with status updates:

Thinking...
Analyzing satellite data...
Processing request...  Executing: list_recent_imagery({'location_name': 'Iceland'})
Processing request...  Executing: create_rgb_composite({...})
Searching imagery...

The spinner cycles through states while the agent processes your query. Once finished, the result is displayed cleanly.


Output Files

All generated images are saved to the outputs/ directory in your working folder:

outputs/
+-- rgb_composite_254.png      # Iceland RGB
+-- rgb_composite_3086.png     # Paris RGB
+-- rgb_composite_7066.png     # Scotland RGB
+-- ndvi_9664.png              # Scotland NDVI
+-- ...

On Windows, the latest image is opened automatically in the default viewer. On macOS and Linux, it uses open and xdg-open respectively.


Supported Models

GeoMind's default model is nvidia/nemotron-3-nano-30b-a3b:free - a free model via OpenRouter. You can switch to any OpenRouter-compatible model that supports function calling:

# Use Claude 3.5 Sonnet
geomind --model "anthropic/claude-3.5-sonnet"

# Use GPT-4
geomind --model "openai/gpt-4"

# Use the free default
geomind

Model Flexibility

Any model available on OpenRouter that supports function calling can be used with GeoMind. Visit the OpenRouter model list to see all available options.