Documentation
Quick start
Aquedra Run sells GIS tools that AI agents call, priced per call, over REST and MCP. Connect a client, or call the API with a key, and every call shows its price before it runs.
Connect over MCP
The server accepts initialize and tools/list without a key; the first tool call starts the sign-in (OAuth) in your client.
Claude
Settings, Connectors, Add custom connector, and paste this URL:
https://r.aquedra.com/mcp/connectorChatGPT
Turn on developer mode, add a connector and use the same URL:
https://r.aquedra.com/mcp/connectorClaude Code
Run this command:
claude mcp add --transport http aquedra-run https://r.aquedra.com/mcpCursor
Add this to your mcp.json:
{
"mcpServers": {
"aquedra-run": {
"url": "https://r.aquedra.com/mcp"
}
}
}Call the REST API
Create an API key in the dashboard and send it as a Bearer token. This call uses dry_run, so it returns the price and charges nothing.
curl
curl -X POST https://r.aquedra.com/v1/geo/measure \
-H "Authorization: Bearer rk_..." \
-H "Content-Type: application/json" \
-d '{"geojson":{"type":"Point","coordinates":[-70.65,-33.45]},"dry_run":true}'Python
import requests
response = requests.post(
"https://r.aquedra.com/v1/geo/measure",
headers={"Authorization": "Bearer rk_..."},
json={"geojson": {"type": "Point", "coordinates": [-70.65, -33.45]}, "dry_run": True},
)
print(response.json())Create an API keyTools and prices
Live prices; they are the same ones /pricing.json lists. Each row is a REST route and an MCP tool.
| Tool | REST route | MCP tool | Price |
|---|---|---|---|
| Coordinate conversion | POST /v1/geo/coordinates | convert_coordinates | US$0.001 |
| GIS measurement | POST /v1/geo/measure | measure_geojson | US$0.001 |
| GIS reprojection | POST /v1/geo/reproject | reproject_geojson | US$0.001 |
| GIS transformation | POST /v1/geo/transform | transform_geojson | US$0.001 |
| GIS file conversion | POST /v1/geo/convert | convert_geo_file | US$0.01 |
Inference
Chat completions on Run's own GPUs, OpenAI-compatible: set base_url to https://r.aquedra.com/v1 in an OpenAI SDK and use your API key. Best effort: the GPUs are shared, so the service can be busy.
Models available now: qwen3-8b, qwen2.5-3b, qwen2.5-coder-7b, qwen2.5-vl-7b.
More