# aaif-goose-goose

- repo: https://github.com/aaif-goose/goose
- commit: ac6222029fe526a4ac15540cb69e2e49cf3a8878
- api style: anthropic
- runs: 2 (None with a reward)
- tasks tried: 1
- models: bedrock/us.anthropic.claude-haiku-4-5-20251001-v1:0

## Results by task

| task | runs | last reward | best reward | last tests |
|---|---|---|---|---|
| [aider_polyglot/polyglot_python_bowling](https://harnessreport.com/tasks/aider_polyglot/polyglot_python_bowling.md) | 2 | 1 |  | 31 passed |

## Runs

| run | task | verifier says | calls | seconds |
|---|---|---|---|---|
| [20260924T154034-aaif-goose-g-polyglot_python_bowling](https://harnessreport.com/runs/20260924T154034-aaif-goose-g-polyglot_python_bowling.md) | aider_polyglot/polyglot_python_bowling | reward 1 · 31 passed | 16 | 79 |
| [20260923T115527-aaif-goose-g-polyglot_python_bowling](https://harnessreport.com/runs/20260923T115527-aaif-goose-g-polyglot_python_bowling.md) | aider_polyglot/polyglot_python_bowling | reward 0 · 31 failed | 8 | 10 |

---
Harness Report runs agent harnesses from their GitHub repos on Harbor tasks and records every model call. Every page is also `.md` and `.json`; index: https://harnessreport.com/llms.txt · MCP: https://harnessreport.com/mcp