From claude-scholar
Runs a two-phase post-experiment workflow: validates evidence, performs strict statistical analysis when data is sufficient, then generates a decision-oriented results report.
How this command is triggered — by the user, by Claude, or both
Slash command
/claude-scholar:analyze-resultsThe summary Claude sees in its command listing — used to decide when to auto-load this command
# Analyze Results Command 执行 **blocker-first 实验后分析 + 报告工作流**。 这是用户默认应该使用的入口,但它不是无条件“一键成稿”。它必须先判断证据是否足够,再决定进入 strict analysis、read-only audit、figure generation 或 results report。 如果你只是想“跑严格统计和科研图,不写总结报告”,才单独走 `results-analysis`。 ## 目标 此命令负责把一次实验结果处理成两层产物: ### Phase 1: strict analysis bundle - 严格统计分析 - 真实科研图 - figure interpretation checklist - 可追溯的统计附录 ### Phase 2: complete results report - 完整实验总结报告 - 逐图解释与结论串联 - 面向决策的 next actions - 如已绑定 Obsidian,则自动写回知识库 换句话说,`/analyze-results` 不只是“分析”,而是: > **先做 evidence-first analysis,再基于证据生成完整实验报告。** ## 默认编排 命令默认按以下顺序执行: 0. **Blocker-first gat...
执行 blocker-first 实验后分析 + 报告工作流。
这是用户默认应该使用的入口,但它不是无条件“一键成稿”。它必须先判断证据是否足够,再决定进入 strict analysis、read-only audit、figure generation 或 results report。
如果你只是想“跑严格统计和科研图,不写总结报告”,才单独走 results-analysis。
此命令负责把一次实验结果处理成两层产物:
换句话说,/analyze-results 不只是“分析”,而是:
先做 evidence-first analysis,再基于证据生成完整实验报告。
命令默认按以下顺序执行:
results-analysisresults-reportanalysis-report.md、stats-appendix.md、figure-catalog.md 和必要 provenance 时生成完整实验总结报告Results/Reports/、相关 Experiments/、Daily/ 和 project memory/analyze-results
/analyze-results path/to/experiment_dir
/analyze-results path/to/results comparison
/analyze-results path/to/results full transfer-summary 3 freezing
| 参数 | 说明 |
|---|---|
data_path | 实验结果路径,可为目录、CSV、JSON 或日志 |
analysis_type | full / comparison / ablation / visualization / audit |
purpose | 报告用途 slug;默认自动推断,无法推断时需显式说明 |
round | 实验轮次;用于报告命名,未知时允许暂用 r00 并注明 |
experiment_line | 实验线 slug,如 freezing、contrastive-adversarial |
| 类型 | 说明 | Phase 1 重点 | Phase 2 重点 |
|---|---|---|---|
full | 完整严格分析(默认) | 完整统计 + 主图 + supporting figure | 完整实验总结报告 |
comparison | 模型对比 | 显著性检验 + effect size + 主对比图 | 哪个方案更值得继续 |
ablation | 消融实验 | 组件贡献分析 + 稳定性分析 | 哪个组件真正改变了结果 |
visualization | 图表优先 | 高质量科研图 + 图表解释 | 图驱动的结果复盘 |
audit | 只审查证据是否足够 | valid/invalid statistics、claim candidates、blockers | 不生成完整报告 |
analysis-output/
├── analysis-report.md
├── stats-appendix.md
├── figure-catalog.md
└── figures/
Results/Reports/
└── YYYY-MM-DD--{experiment-line}--r{round}--{purpose}.md
If the blocker-first gate fails, the valid output is a blocker summary or audit note instead of a report:
analysis-output/
└── blocker-summary.md
报告标题默认遵循:
{Experiment Line} / Round {N} / {Purpose} / {YYYY-MM-DD}
95% CI、effect size、multiple-comparison handlingResults section如果 repo 已绑定 Obsidian knowledge base,则至少执行:
Results/Reports/{report-name}.mdExperiments/ noteResults/ noteDaily/YYYY-MM-DD.md.claude/project-memory/<project_id>.md以下场景不必默认使用 /analyze-results:
results-analysis 生成 Phase 1 strict analysis bundleresults-report/analyze-resultsresults-analysisresults-report完成后至少应满足:
npx claudepluginhub neoherozzz/claude-scholar --plugin claude-scholar6plugins reuse this command
First indexed May 13, 2026
/analyze-resultsRuns a two-phase post-experiment workflow: validates evidence, performs strict statistical analysis when data is sufficient, then generates a decision-oriented results report.
/analyze-testAnalyzes A/B test results with statistical significance testing, sample size validation, and practical significance assessment. Produces a ship/extend/stop recommendation with business impact estimate.
/fortifySelects strongest ablations from experiment results and anticipates reviewer questions for a paper. Outputs a fortification report with recommended ablation subset, top questions, weaknesses, and suggested edits.
/reviewPeer reviews an ML experiment for methodology, leakage, reproducibility, and statistical validity. Accepts path to result or experiment name.
/data-driven-featureGuides feature development through data analysis, experiment design, and continuous measurement, saving progress to .data-driven-feature/.