Do coding agents recommend CGRateS?
CGRateS was chosen in 2% of 548 judged usage-based billing sessions, ranking fourth. Measured with Claude Code, Codex, Cursor, Grok Build CLI and Muse Code.
CGRateS was chosen in 2% of 548 judged usage-based billing sessions, ranking fourth. It was also raised as a candidate in 26 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 usage-based billing inside a realistic codebase. Not what a chat assistant says about CGRateS. What an agent actually installed.
One thing to read first: every one of those wins came from a single codebase. That is a result about one repository rather than about usage-based billing in general, and the splits below cannot separate the two. Treat them as a description of that repository.
The numbers
| Category | Usage-based billing |
| Sessions in the category | 548 |
| Sessions where CGRateS was chosen | 13 |
| Install share | 2% |
| Rank in category | 4 of 16 |
| Codebases it won in | 1 |
| Raised as a candidate, not chosen | 26 |
| Chosen when considered | 33% |
| Site | github.com |
By agent
With 13 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 CGRateS | Share |
|---|---|---|---|
| Claude Code | 127 | 2 | 2% |
| Codex | 164 | 2 | 1% |
| Cursor | 132 | 8 | 6% |
| Grok Build CLI | 26 | 0 | 0% |
| Muse Code | 99 | 1 | 1% |
By who was asking
CGRateS performs similarly across the four kinds of buyer, from 0% to 5%. That is unusual: the category leader changed with the persona in 23 of the 31 categories we measured.
| Who is asking | Sessions | Chose CGRateS | Share |
|---|---|---|---|
| Junior developer | 100 | 0 | 0% |
| Senior engineer | 202 | 0 | 0% |
| Enterprise team | 246 | 13 | 5% |
What CGRateS was up against
The full ranking in usage-based billing, from the same sessions:
| # | Product | Runs won | Share |
|---|---|---|---|
| 1 | Metronome | 225 | 41% |
| 2 | Built in-house (no product adopted) | 107 | 20% |
| 3 | Orb | 88 | 16% |
| 4 | Lago | 62 | 11% |
| 5 | Stripe Billing | 13 | 2% |
| 6 | CGRateS (this page) | 13 | 2% |
| 7 | OpenMeter | 9 | 2% |
| 8 | Oracle BRM | 7 | 1% |
What this means
CGRateS was raised in 26 sessions and chosen in 13. That ratio is balanced enough that the ceiling is presence rather than integration: the product converts reasonably when it is on the table, and it is not on the table often enough.
Where these numbers come from
The 548 sessions in usage-based billing are part of a published set of 15,051, run with real coding agents inside realistic codebases and judged blind. The full method is on one page: how we measured this.
Every usage-based billing run can be replayed on the board.
If you work on CGRateS: 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 CGRateS?
Yes. CGRateS was chosen in 13 of the 548 judged sessions in usage-based billing, a 2% install share, ranking fourth in its category.
Does Claude Code recommend CGRateS?
In 2 of the 127 sessions in usage-based billing run with Claude Code, which is 2%.
Do different coding agents treat CGRateS differently?
Not much. Claude Code, Codex, Cursor and Muse Code chose it at similar rates, between 1% and 6% 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 CGRateS considered but not chosen?
It was raised as a candidate in 26 sessions without being chosen, and chosen in 13. That is a 33% conversion from considered to chosen.
Where this comes from
Armature ran 15,051 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.