菜单
Decompose research ideas into atomic, self-contained concepts with bidirectional math-code mapping. For each concept, extract the math formula from papers and find code implementations. Use for complex system papers requiring formal grounding.
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
Conduct systematic academic literature reviews in 6 phases, producing structured notes, a curated paper database, and a synthesized final report. Output is organized by phase for clarity.
Explore and analyze GitHub repositories related to a research topic. Reads deep-research output, discovers repos from multiple sources, deeply analyzes code, and produces integration blueprints.
Manage BibTeX citations for LaTeX papers. Harvest missing citations from a draft using Semantic Scholar, validate cite keys against .bib files, deduplicate entries, and format bibliography. Use when working with references, BibTeX, or citations.
Generate statistical analysis code with 4-round review. Select appropriate statistical tests, interpret results, and produce analysis reports with p-values, effect sizes, and confidence intervals. Use when analyzing experimental data for a paper.
Write ML experiment code with iterative improvement. Generate training/evaluation pipelines, debug errors, and optimize results through code reflection. Use when implementing experiments for a research paper.
Generate publication-quality scientific figures using matplotlib/seaborn with a three-phase pipeline (query expansion, code generation with execution, VLM visual feedback). Handles bar charts, line plots, heatmaps, training curves, ablation plots, and more. Use when the user needs figures, plots, or visualizations for a paper.
| name | atomic-decomposition |
| description | Decompose research ideas into atomic, self-contained concepts with bidirectional math-code mapping. For each concept, extract the math formula from papers and find code implementations. Use for complex system papers requiring formal grounding. |
| argument-hint | ["idea-or-paper"] |
Decompose research ideas into atomic concepts with math formula <-> code implementation mapping.
$0 — Research idea, paper, or method description~/.claude/skills/atomic-decomposition/references/decomposition-prompts.mdAnalyze the research idea and decompose into atomic, self-contained concepts:
{
"definition": "Kernelized Gumbel-Softmax Operator",
"math_formula": "Z = \\text{softmax}((\\log \\pi + g) / \\tau), g \\sim \\text{Gumbel}(0,1)",
"code_implementation": "def gumbel_softmax(logits, tau=1.0): ...",
"reference_papers": ["Paper Title 1"],
"reference_codebases": ["github_user/repo_name"],
"assumptions": ["Differentiable relaxation of discrete sampling"],
"connections": ["Used in Component X of the proposed method"]
}