The dataset
Check the repository commits, languages, tool versions, settings, and evaluation date before comparing results.
ASTER / THE BENCHMARK LAB
A better review
starts with
the evidence.
Explore reported detection rates, the bugs behind the numbers, and what each evaluation can tell you.
Visible sourcesReported results
BUG DETECTION RATEREPORTED RESULTS
Aster / reported result
Competitor source: Macroscope’s published benchmark54.1% detection, reported by Aster. The supporting report and test details will be linked here when supplied.
54.1%ASTER-REPORTED DETECTION RATE
A result with room for the detail.
Aster reports 54.1% bug detection. The supporting report link, test date, dataset, and bug counts have not yet been supplied. We’ll add those details here alongside the result.
Supporting report
Link to follow
Link to follow
THE REPORTED COMPARISON
Every score.
With its context.
Aster’s reported result appears alongside historical competitor results from Macroscope’s 2025 evaluation.
| Tool | Detection rate | Bugs found / evaluated | Source |
|---|---|---|---|
| Aster | 54.1% | Not yet supplied | Aster · report to follow |
| Macroscope | 48.3% | 57 / 118 | Macroscope · 2025 study |
| CodeRabbit | 45.8% | 54 / 118 | Macroscope · 2025 study |
| Cursor Bugbot | 42.4% | 50 / 118 | Macroscope · 2025 study |
| Greptile | 23.6% | 17 / 72 | Macroscope · 2025 study |
| Graphite Diamond | 18.3% | 21 / 115 | Macroscope · 2025 study |
Competitor source: Macroscope’s September 2025 evaluation, tested August 25–September 14, 2025. Completion counts differ by tool. Aster’s test conditions have not yet been supplied, so the figures do not establish a like-for-like performance advantage or a current independent ranking.
WHAT MAKES A SCORE USEFUL
A clear test.
A result you can inspect.
Read a benchmark through its inputs, evidence, and limits.
The evidence
Findings should map to known bugs. A detection rate alone does not describe incorrect or noisy suggestions.
The denominator
Failed and skipped cases matter. Check the number of evaluated bugs for each tool, alongside the percentage.
The source
Macroscope publishes the competitor study linked here. Aster’s supporting report will be added when available.
Runtime bugs and security coverage measure different things.
The cited study focuses on known runtime bugs across eight languages. Aster’s current source audit checks credential patterns, JavaScript/TypeScript syntax, and selected risky-code patterns. An overall bug-detection score is not a security guarantee. Read Aster’s current scope.
START WITH YOUR OWN CODE
Your repository.
Your first closer look.
Get the findings, source locations, and next steps from your first audit. Every finding is included.