Python Code Tester
ReviewAudited by ClawScan on May 10, 2026.
Overview
This skill is openly meant to test and fix Python code, but it can run generated tests and update real project files without a clear approval step.
Use this skill only in a version-controlled or disposable workspace. Review generated tests and diffs before allowing repaired code to be copied back into real project files, and install its dependencies in an isolated Python environment.
Findings (5)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The agent could change working project files after a failed test, potentially introducing incorrect fixes or overwriting local work.
The documented default workflow includes repairing code and updating the real project files, but does not require a user-approved diff, confirmation, branch, backup, or rollback before mutating the project.
若测试失败，分析错误原因 - 尝试修复代码 ... 将修复后的代码更新到项目实际文件中
Only use this in a version-controlled or backed-up workspace, and require explicit user review of diffs before syncing any repaired code into the actual project.
The skill may inspect more local Python files than the user expected while trying to find the target code.
The helper script recursively searches Python files under the computed project root. This is broadly aligned with code testing, but it is wider than the SKILL.md wording that search should be limited to a single class or function.
py_files = list(project_root.glob("**/*.py"))Point it at a narrow project/workspace and verify the files it identifies before allowing testing or repair.
Running tests can execute project code and any side effects those tests trigger.
The skill is designed to generate and run test scripts locally. This is expected for its purpose, but it means generated code and project code may be executed.
编写测试脚本放在 `scripts/` 目录 - 运行测试，结果保存到 `scripts/log/` 目录
Run it in a controlled environment and avoid using it on codebases where test execution could affect production data or services.
Dependency versions may vary over time, which can affect reproducibility or introduce dependency-chain risk.
The setup script installs Python dependencies, while requirements.txt uses non-pinned lower-bound ranges such as pytest>=7.0.0, numpy>=1.21.0, and pandas>=1.3.0.
pip install -r "$SKILL_DIR/scripts/requirements.txt"
Review dependencies before installation and consider pinning exact versions or installing in a virtual environment.
Project code copies, test data, and error logs may remain on disk after the task completes.
The skill persists generated test data, logs, error output, and repaired versions in local directories. This is disclosed and purpose-aligned, but it may retain project-related information.
必须将测试数据保存到 `references/` ... 成功修复的版本保存到 `scripts/release/v{n}/` ... 所有测试结果和错误信息都要记录到log目录Inspect and clean the references, scripts/log, and scripts/release directories if they may contain sensitive project details.