- VERIFIED
- DETECTED
- REFINED
- HUMAN
Short answer
How do I check if my AI's spatial math is right?
Your agent uses the calculate_distance tool to run spherical trigonometry on specific coordinates. This provides a ground-truth distance to compare against the agent's internal estimates. You get a mathematical baseline to catch hallucinations in spatial reasoning.
The outcomes of validation
Where your results land
Running this task produces specific results for your spatial workflows.
VERIFIED
Accurate distance checks
The agent uses the tool to get a precise measurement. You receive a confirmed distance to use in your final output.
DETECTED
Caught spatial errors
When the agent's estimate differs from the tool result, you'll see the discrepancy immediately. This flags reasoning failures.
REFINED
Improved agent logic
You can feed the correct distance back into the chat. This helps the agent correct its internal spatial model.
HUMAN
Manual review needed
If the distance is zero or coordinates are invalid, the agent flags it. You decide how to handle these edge cases.
The workflow
What your AI does when the request arrives.
The agent moves from raw coordinates to a verified mathematical truth.
Identify coordinates
The agent extracts latitude and longitude pairs from your prompt or data source.
calculate_distanceExecute math
The agent calls the engine to perform the Haversine formula on those points.
calculate_distanceCompare results
The agent compares the tool's output against its own predicted distance.
calculate_distanceReport findings
The agent tells you if the spatial reasoning was correct or if it needs to adjust.
calculate_distance
Try it
Copy these to start.
Use these prompts to begin testing your agent's spatial accuracy.
Starting points
Replace the locations and coordinates with your specific data.
Haversine Engine Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_9HY5DGRNuTnVYKh6pT2quv71WB3YfM9cSRfriCCA/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — Haversine Engine capabilities are ready to use.
{
"mcpServers": {
"haversine-distance-engine-mcp": {
"url": "https://edge.vinkius.com/vk_preview_9HY5DGRNuTnVYKh6pT2quv71WB3YfM9cSRfriCCA/mcp"
}
}
}Calculate the distance between the Empire State Building and Times Square to see if your estimate is correct.
Check the distance between 40.7128° N, 74.0060° W and 34.0522° N, 118.2437° W.
I think these two points in London are 5 miles apart. Use the tool to verify this.
Compare the distance between the coordinates in this CSV row to my previous spatial estimate.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Haversine Distance Engine once, then ask.
One connection gives your agent access to the math engine. Your credentials stay encrypted and secure.
Connect Haversine Distance Engine to your AIFAQ
How this task behaves
- 01
Can the AI change the coordinates itself?
No. The agent can only use the coordinates you provide or those it extracts from your text.
- 02
Does the tool calculate flight paths?
No. It only calculates the great-circle distance between two points on a sphere.
- 03
Can the AI fix my spreadsheet automatically?
No. The tool only provides the distance calculation; you must manage your own files.
- 04
What happens if I provide invalid coordinates?
The tool will return an error, and the agent will report that the calculation failed.
- 05
Can the agent use this to navigate a vehicle?
No. This is a mathematical verification tool, not a real-time navigation system.
More questions about Haversine Distance Engine? The Connector page answers them. See everything the Haversine Distance Engine Connector can do
Connect Haversine Engine to Claude, Cursor, ChatGPT & more
