shinka-inspect
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChineseShinka Inspect Skill
Shinka Inspect Skill
Extract the strongest programs from a Shinka run and package them into a context file that coding agents can load directly.
提取Shinka运行过程中表现最优的程序,并将其打包为编码Agent可直接加载的上下文文件。
When to Use
适用场景
Use this skill when:
- A run already produced a results directory and SQLite database
- You want to inspect top-performing programs before launching the next batch
- You want a compact context artifact instead of manually browsing the DB
Do not use this skill when:
- You still need to scaffold a task ()
shinka-setup - You need to run evolution batches ()
shinka-run
在以下场景使用该技能:
- 某次运行已生成结果目录和SQLite数据库
- 你希望在启动下一批次前检查表现最优的程序
- 你需要一个简洁的上下文工件,而非手动浏览数据库
请勿在以下场景使用:
- 你仍需要搭建任务框架()
shinka-setup - 你需要运行进化批次()
shinka-run
What it does
功能说明
- Uses to read program records
shinka.utils.load_programs_to_df - Ranks programs by
combined_score - Selects top-correct programs (fallback to top-
koverall if no correct rows)k - Writes one Markdown bundle with metadata, ranking table, feedback, and code snippets
- 使用读取程序记录
shinka.utils.load_programs_to_df - 按对程序进行排名
combined_score - 选择排名前k的正确程序(如果没有正确行,则回退到整体排名前k的程序)
- 生成一个包含元数据、排名表、反馈和代码片段的Markdown包
Workflow
工作流程
- Confirm run artifacts exist
bash
ls -la <results_dir>- Generate context bundle
bash
python skills/shinka-inspect/scripts/inspect_best_programs.py \
--results-dir <results_dir> \
--k 5- Optional tuning knobs
bash
python skills/shinka-inspect/scripts/inspect_best_programs.py \
--results-dir <results_dir> \
--k 8 \
--max-code-chars 5000 \
--min-generation 10 \
--out <results_dir>/inspect/top_programs.md- Load output into agent context
- Default output path:
<results_dir>/shinka_inspect_context.md - Use it as the context artifact for next-step mutation planning
- 确认运行工件存在
bash
ls -la <results_dir>- 生成上下文包
bash
python skills/shinka-inspect/scripts/inspect_best_programs.py \
--results-dir <results_dir> \
--k 5- 可选调优参数
bash
python skills/shinka-inspect/scripts/inspect_best_programs.py \
--results-dir <results_dir> \
--k 8 \
--max-code-chars 5000 \
--min-generation 10 \
--out <results_dir>/inspect/top_programs.md- 将输出加载到Agent上下文
- 默认输出路径:
<results_dir>/shinka_inspect_context.md - 将其用作下一步突变规划的上下文工件
CLI Arguments
CLI参数
- : Path to run directory (or direct DB file path)
--results-dir - : Number of programs to include (default
--k)5 - : Output markdown path (default under results dir)
--out - : Per-program code truncation cap (default
--max-code-chars)4000 - : Optional lower bound on generation
--min-generation - /
--include-feedback: Include--no-include-feedbackblockstext_feedback
- :运行目录路径(或直接指定数据库文件路径)
--results-dir - :要包含的程序数量(默认值为
--k)5 - :Markdown输出路径(默认位于结果目录下)
--out - :单个程序代码的截断上限(默认值为
--max-code-chars)4000 - :可选的生成次数下限
--min-generation - /
--include-feedback:是否包含--no-include-feedback块text_feedback
Notes
注意事项
- Ranking metric is .
combined_score - If no correct rows exist, script falls back to top-score rows and labels fallback in output.
- Script is read-only for run artifacts (writes only the markdown bundle).
- 排名指标为。
combined_score - 如果没有正确行,脚本会回退到最高分的行,并在输出中标记回退状态。
- 脚本对运行工件仅为只读操作(仅写入Markdown包)。