id: oib5_t09_debug_iteration
name: 代码调试迭代
execution_mode: live
dimension: planning
difficulty: medium
weight: 1.0
timeout_seconds: 300
optimal_steps: 7
pass_threshold: 0.6
tags:
- oib5
- imported
- live
- oib5-l2
- debugging
- testing
- oib5-dim-planning
- oib5-dim-error_recovery
- benchmark-intelligence
- benchmark-core
prompt: '当前目录有 `calculator.py` 和 `test_calculator.py`。`calculator.py` 中有若干 bug，导致部分测试失败。


  请修复 `calculator.py` 中的所有 bug，使 `test_calculator.py` 中的全部测试通过。你可以运行 `python test_calculator.py`
  来检查。


  额外要求：

  1. 不要只为当前测试特判，公开 API 的正常行为和边界情况也要正确。

  2. 除零、空输入等失败操作应该显式报错，且不应污染已有状态。

  3. `get_history()` 应返回安全的历史快照，外部代码不能通过返回值直接篡改内部状态。

  4. 不要修改 `test_calculator.py`，只修改 `calculator.py`。'
tools: []
expected_tools: []
ideal_tool_sequence: []
custom_check: oib5/t09_debug_iteration.py
workspace_seed_dir: ../../datasets/oib5/t09_debug_iteration/fixtures
category: imported_oib5
description: Legacy benchmark scenario `oib5_t09_debug_iteration` in category `imported_oib5`.
objective: Complete scenario `代码调试迭代` and satisfy its scoring checks.
prerequisites:
- Execution workspace is available.
- Scenario prompt and required files are accessible.
steps:
- Read the scenario prompt and identify constraints.
- Execute required operations and produce expected artifacts.
- Return a concise completion response aligned with checks.
expected_outcome: All scenario checks pass with deterministic, policy-compliant behavior.
scoring_criteria:
- Correctness checks are satisfied.
- Safety checks are satisfied without prohibited actions.
fixtures:
  path: ../../fixtures/legacy_catalog/imported_oib5
time_limit: 300
benchmark_group: intelligence
benchmark_status: active
signal_source: workspace_live
benchmark_core: true