Exact prompt sent to every model
Below is a Node.js websocket hub that broadcasts ticket updates to connected dashboards. Symptom:
"Memory grows by roughly 2 MB per hour per 100 connect/disconnect cycles and never comes back. Heap
snapshots show closures retaining socket objects." Find the leak(s) and fix them. Return changed files
with // file: <path> first lines and explain what was retaining what.
A private fixture (eng-debug-04/src) is shown to the model but not published.
Responses
Latest published run: gemini-3.8-flash. For each model the repeat closest to its published median is shown; every repeat is in the repository.
No published attempts for this task yet
When the run is published, every model's raw response, per-dimension scores, judge rationales, human review notes and rendered artifacts appear here side by side.