nanoresearch-planning

从选定假设生成详细实验蓝图，指定数据集、基线与评估指标。

从工作目录的 outline.yaml 自动读取待调研条目，为每个或每批 item 启动独立 agent 并行深度调研。适合需要把大纲拆成可并行执行的调研任务、支持断点续传和分批审批的场景。运行时会跳过已完成项、分批执行并关闭 task output，便于后台大规模采集而不干扰主流程。

# Planning Skill

## Purpose
Take the selected hypothesis from ideation and produce a detailed experiment blueprint specifying datasets, baselines, evaluation metrics, and ablation groups.

## Tools Required
None. This skill operates entirely through LLM reasoning over the ideation output.

## Input
- `ideation_output`: Path to `papers/ideation_output.json` produced by the ideation skill

## Process
1. Parse the selected hypothesis and supporting literature from the ideation output
2. Identify candidate datasets that are publicly available and appropriate for validating the hypothesis
3. Select 2-4 baseline methods from the surveyed literature for comparison
4. Define primary and secondary e