# nanocoai-nanoclaw

- repo: https://github.com/nanocoai/nanoclaw
- commit: 143db6c907c652773a536c7c9e96269fdad0a4a4
- 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 | 0 |  | 14 failed, 17 passed |

## Runs

| run | task | verifier says | calls | seconds |
|---|---|---|---|---|
| [20260924T154034-nanocoai-nan-polyglot_python_bowling](https://harnessreport.com/runs/20260924T154034-nanocoai-nan-polyglot_python_bowling.md) | aider_polyglot/polyglot_python_bowling | reward 0 · 14 failed, 17 passed | 4 | 22 |
| [20260923T140508-nanocoai-nan-polyglot_python_bowling](https://harnessreport.com/runs/20260923T140508-nanocoai-nan-polyglot_python_bowling.md) | aider_polyglot/polyglot_python_bowling | reward 0 · 31 failed | 1 | 5 |

---
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