Do coding agents recommend Umami?
Umami was chosen in 1% of 598 judged product analytics sessions, ranking eighth. Measured with Claude Code, Codex, Cursor, Grok Build CLI and Muse Code.
Umami was chosen in 1% of 598 judged product analytics sessions, ranking eighth. It was also raised as a candidate in 91 further sessions without being chosen.
This page reports what happened when Claude Code, Codex, Cursor, Grok Build CLI and Muse Code had to solve a problem in product analytics inside a realistic codebase. Not what a chat assistant says about Umami. What an agent actually installed.
The numbers
| Category | Product analytics |
| Sessions in the category | 598 |
| Sessions where Umami was chosen | 8 |
| Install share | 1% |
| Rank in category | 8 of 20 |
| Codebases it won in | 2 |
| Raised as a candidate, not chosen | 91 |
| Chosen when considered | 8% |
| Site | umami.is |
By agent
With 8 wins spread across 5 agents, the rates below are small numbers and a difference between them is not yet a finding. They are here because the direction is worth knowing, not because the gap is established.
| Agent | Sessions | Chose Umami | Share |
|---|---|---|---|
| Claude Code | 150 | 0 | 0% |
| Codex | 180 | 7 | 4% |
| Cursor | 149 | 1 | 1% |
| Grok Build CLI | 29 | 0 | 0% |
| Muse Code | 90 | 0 | 0% |
By who was asking
Umami performs similarly across the four kinds of buyer, from 0% to 3%. That is unusual: the category leader changed with the persona in 23 of the 31 categories we measured.
| Who is asking | Sessions | Chose Umami | Share |
|---|---|---|---|
| Vibe coder | 60 | 1 | 2% |
| Junior developer | 240 | 0 | 0% |
| Senior engineer | 238 | 7 | 3% |
| Enterprise team | 60 | 0 | 0% |
What Umami was up against
The full ranking in product analytics, from the same sessions:
| # | Product | Runs won | Share |
|---|---|---|---|
| 1 | PostHog | 332 | 56% |
| 2 | Built in-house (no product adopted) | 82 | 14% |
| 3 | Metabase | 37 | 6% |
| 4 | Vercel Analytics | 27 | 5% |
| 5 | Mixpanel | 19 | 3% |
| 6 | Amplitude | 19 | 3% |
| 7 | Segment | 18 | 3% |
| 8 | Looker Studio | 12 | 2% |
What this means
This is an integration problem, not a presence problem. Agents raised Umami in 91 sessions and chose it in 8, 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 598 sessions in product analytics are part of a published set of 15,000, run with real coding agents inside realistic codebases and judged blind. The full method is on one page: how we measured this.
Every product analytics run can be replayed on the board.
If you work on Umami: 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 Umami?
Yes. Umami was chosen in 8 of the 598 judged sessions in product analytics, a 1% install share, ranking eighth in its category.
Does Claude Code recommend Umami?
In 0 of the 150 sessions in product analytics run with Claude Code, which is 0%.
Do different coding agents treat Umami differently?
Not much. Claude Code, Codex, Cursor and Muse Code chose it at similar rates, between 0% and 4% of their runs.
How was this measured?
Real coding agents at pinned versions were run in sandboxes inside 92 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 Umami considered but not chosen?
It was raised as a candidate in 91 sessions without being chosen, and chosen in 8. That is a 8% conversion from considered to chosen.
Where this comes from
Armature ran 15,000 judged sessions with Claude Code, Codex, Cursor, Grok Build CLI and Muse Code inside 92 realistic codebases, and published every run. The numbers on this page come from that work.