菜单
This skill should be used when the user asks to analyze, reproduce, or run a computational math repository. Use when inspecting source code, detecting algorithms, planning runs, executing experiments, or collecting results from optimization code.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Route AI4Math computational mathematics tasks to the appropriate bundled skill package.
Use when reproducing, deploying, diagnosing, tuning, visualizing, or reporting computational math research code with a coding agent
This skill should be used when the user asks to search for external algorithms, find related implementations, discover alternative approaches, or look up papers and code for computational math optimization methods.
This skill should be used when the user asks to tune parameters, run hyperparameter search, optimize algorithm settings, or find the best configuration for a computational math method.
Use when classifying a computational math repository or task into a broad domain before choosing specialist Skills, references, runtimes, and validation evidence.
Use when a user starts an end-to-end computational math research-code reproduction, deployment, tuning, visualization, or reporting workflow with a coding agent.
| name | repo-reproduction-skill |
| description | This skill should be used when the user asks to analyze, reproduce, or run a computational math repository. Use when inspecting source code, detecting algorithms, planning runs, executing experiments, or collecting results from optimization code. |
Analyzes a computational math repository, generates run plans, executes them, and collects results.
For end-to-end computational math research-code reproduction workflows, this Skill should be selected by computational_math_reproduction_workflow_skill rather than used as the first entrypoint.
Analyzes source code directory. Input: --source <path>, --out <path>. Writes analysis summary to plan.md fields (source path, language, entrypoints, detected algorithms, risks) — does not write repo_analysis.json by default.
python -m skills.repo_reproduction_skill.scripts.repo_analyzer --source /path/to/repo --out /path/to/output
Generates candidate run plans from the analysis. Input: --analysis <dict or json>, --out <path>. Outputs compact plan fields for plan.md — does not write run_plan.json by default.
python -m skills.repo_reproduction_skill.scripts.run_planner --analysis /path/to/analysis.json --out /path/to/output
Executes an approved run plan. Input: --command <str>, --out <path>, --timeout <int>. Writes stdout/stderr to logs/run.log and returns exit code. Does not write run_plan.json or execution_log.jsonl by default.
python -m skills.repo_reproduction_skill.scripts.executor --command "python main.py" --out /path/to/output --timeout 300
Scans executed output for result files and extracts metrics. Input: --source, --out. Writes results to results/ — does not write collected_results.json by default.
python -m skills.repo_reproduction_skill.scripts.result_collector --source /path/to/repo --out /path/to/output
repo_analysis.json by default. Summarize analysis in conversation and in plan.md fields.run_plan.json by default. Use compact plan fields.logs/run.log.results/.figures/.patches/.repo_analyzer to understand the repository.plan.md fields.executor with the approved command.result_collector to gather results under results/.