A shorter route
through the code.
Measured on the oldest 30 held-out Django fixes with claude-sonnet-5-5. Numbers are means per issue. Accuracy is about the same with and without the tool (top-1 function 0.43 vs 0.40, one issue). The tool cut input tokens by 16 percent, or 27 percent when the agent is told to call it first.
Without RepoCortex
search, read, repeat- Input tokens
- 63,885
- Turns
- 9.6
- Files read
- 2.8
With RepoCortex
-16% input tokens- Input tokens
- 53,888
- Turns
- 8.3
- Files read
- 2.4
Told to call it first
-27% input tokens- Input tokens
- 46,361
- Turns
- 7.5
- Files read
- 2.2
The eval record
Means per issue| Metric | Without | With | Call it first | locate_bug alone |
|---|---|---|---|---|
| Top-1 function correct | 0.40 | 0.43 | 0.43 | 0.47 |
| Top-1 file correct | 0.63 | 0.63 | 0.63 | 0.67 |
| Input tokens per issue | 63,885 | 53,888 (-16%) | 46,361 (-27%) | 0 |
| Tool calls per issue | 11.3 | 9.8 | 9.1 | 1 |
| Files opened per issue | 2.8 | 2.4 | 2.2 | 0 |
| Cost per issue | $0.063 | $0.054 (-15%) | $0.049 (-22%) | $0 |
Accuracy is within noise across arms: one issue out of 30. locate_bug alone names the right function first 0.47 of the time against 0.40 for the agent searching by itself, in under a second and with no model call.
One run, #36277 Add GeneratedField(db_persist=False) support on PostgreSQL 18+
Measured tracelocate_bug
{
"report": "Add GeneratedField(db_persist=False) support on PostgreSQL 18+",
"k": 10
}Measured run on this ticket. With locate_bug the agent made 21 tool calls over 13 turns, used 137,942 input tokens, and named the gold function DatabaseFeatures.is_postgresql_18 second. Without it: 23 tool calls, 15 turns, 166,134 input tokens, gold named third.