From the experiment

Do coding agents recommend MapLibre?

MapLibre was chosen in 7% of 319 judged maps sessions, ranking fifth. Measured with Claude Code, Codex and Cursor.

Published September 11, 2026 Read as Markdown

MapLibre was chosen in 7% of 319 judged maps sessions, ranking fifth. It was also raised as a candidate in 176 further sessions without being chosen.

This page reports what happened when Claude Code, Codex and Cursor had to solve a problem in maps inside a realistic codebase. Not what a chat assistant says about MapLibre. What an agent actually installed.

The numbers

CategoryMaps
Sessions in the category319
Sessions where MapLibre was chosen21
Install share7%
Rank in category5 of 32
Codebases it won in7
Raised as a candidate, not chosen176
Chosen when considered11%
Sitemaplibre.org

By agent

The three agents agree closely on MapLibre, choosing it at rates within 5 points of each other.

AgentSessionsChose MapLibreShare
Claude Code108109%
Codex11176%
Cursor10044%

By who was asking

MapLibre does much better with one kind of buyer than another. It won 36% of sessions asked as enterprise team and 2% of those asked as vibe coder.

Who is askingSessionsChose MapLibreShare
Vibe coder6412%
Junior developer14743%
Senior engineer7234%
Enterprise team361336%

What MapLibre was up against

The full ranking in maps, from the same sessions:

#ProductRuns wonShare
1Google Maps5216%
2Leaflet5216%
3Apple Maps (MapKit)4614%
4Mapbox258%
5MapLibre (this page)217%
6react-native-maps196%
7Azure Maps186%
8Built in-house (no product adopted)165%

What this means

This is an integration problem, not a presence problem. Agents raised MapLibre in 176 sessions and chose it in 21, so it reaches the shortlist and then loses. Something at the last step is costing the session, and in our data that is usually a quickstart that does not run when pasted, documentation describing an interface that changed, or a package name that does not match the product name.

That is the cheaper of the two problems to have. The reason is written down in each losing transcript.

Where these numbers come from

The 319 sessions in maps are part of a published set of 6,458, run with real coding agents inside realistic codebases and judged blind. The full method is on one page: how we measured this.

Every maps run can be replayed on the board.

If you work on MapLibre: the judge recorded a reason for every session where it was raised and passed over. Those reasons are in the transcripts.

<!-- generated by scripts/write-data-pages.mjs -->

Common questions

Do coding agents recommend MapLibre?

Yes. MapLibre was chosen in 21 of the 319 judged sessions in maps, a 7% install share, ranking fifth in its category.

Does Claude Code recommend MapLibre?

In 10 of the 108 sessions in maps run with Claude Code, which is 9%.

Do different coding agents treat MapLibre differently?

Not much. The three agents chose it at similar rates, between 4% and 9% of their runs.

How was this measured?

Real coding agents at pinned versions were run in sandboxes inside 63 realistic codebases and asked to solve real tasks. A simulated project owner approved or questioned each recommendation before any code was written, and a judge from a model family that builds none of the agents read every session blind.

How often is MapLibre considered but not chosen?

It was raised as a candidate in 176 sessions without being chosen, and chosen in 21. That is a 11% conversion from considered to chosen.

Where this comes from

Armature ran 6,458 judged sessions with Claude Code, Codex and Cursor inside 63 realistic codebases, and published every run. The numbers on this page come from that work.

Read next

All library pages