# Run 20260923T152255-tarruda-neoa-polyglot_python_bowling - harness: [tarruda-neoagent](https://harnessreport.com/harnesses/tarruda-neoagent.md) @ 69dc9a11992f - task: [aider_polyglot/polyglot_python_bowling](https://harnessreport.com/tasks/aider_polyglot/polyglot_python_bowling.md) - model: bedrock/us.anthropic.claude-haiku-4-5-20251001-v1:0 - verifier says: reward 0 · 3 failed, 28 passed - model calls: 24 tokens in/out: 248579/14465 - agent seconds: 98 - started: 2026-09-23T15:26:01 - full bundle (calls, logs, files): https://harnessreport.com/api/run/20260923T152255-tarruda-neoa-polyglot_python_bowling ## Failed tests - test_bonus_roll_for_a_spare_in_the_last_frame_must_be_rolled_before_score_can_be_calculated - test_bonus_rolls_for_a_strike_in_the_last_frame_must_be_rolled_before_score_can_be_calculated - test_both_bonus_rolls_for_a_strike_in_the_last_frame_must_be_rolled_before_score_can_be_calculated Last agent action: `shell: cd /app && python3 << 'EOF' from bowling import BowlingGame print("=" * 60) print("BOWLING GAME SCORING IMPLEMENTATION -…` --- 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