This space is very active. I work at the French mapping agency, and we're currently building MCPs to work with our data.

See: https://github.com/ignfab/geocontext (French) Beta MCP instance: https://geollm.beta.ign.fr/geocontext/mcp

Unrelated, but also take a look at the nice high-density LiDAR point data we have! https://visionneuse-lidarhd.ign.fr/?px=4441970.281583222&py=...

>4x improvement on geospatial tasks with map in the loop.

The graph shows a baseline 2% task success rate improving to to 8% task success rate, but the evals section details 100% success rates across the board.

I'm not sure what the effectiveness of this skill is from the readme. Is it 8% success, or 100% success?

Question from an outsider: Who is paying for tools like this? The examples shown on the website (e.g. all streets in Nevada) look nice, but what are those analyses actually used for? I am pretty sure it is not only about having pretty maps but their has to be a business value I don’t see right now.

For the maps-in-loop part, what does the agent actually read back from Dekart each round to catch the geometry errors?

If I see another skill or markdown on hackernews I might just consider leaving the platform. What even is the point of sharing markdowns...

Either LLMs will be so good in a few months this will be redundant.

Or it won't be and LLMs are a dead end and there are better ways to build with LLMs

[dead]

[dead]

[dead]

[flagged]