Lambenthan
User经管实证研究的 AI 知识库 — 从文献阅读到 Stata 执行，一条流水线串到底。基于 Karpathy 的 LLM-Wiki 理念，按实证研究 10 类实体（变量 / 数据集 / 模型 / 机制 / 假设 / 识别策略 / 稳健性 / 异质性 / 表格 / 论文）定制
Categories
Indexed Skills (29)
edit
根据用户要求增删 raw sources 或更新 wiki 内容
daily-arxiv
Daily arXiv pull, relevance filtering, auto-ingest of high-priority papers, and SOTA update detection
discover
Build a ranked shortlist of candidate papers (anchor-driven, topic-driven, or derived from current wiki state) that the user — or an upstream skill — may decide to feed into `/ingest`. Use whenever the user asks "what should I read next", "find papers similar to this one", "recommend related work", "what's around this topic", or whenever `/ingest` is invoked with `--discover`. Does not ingest; only proposes.
empirical-ingest
将一篇经管实证论文摄取为实证研究 wiki：论文卡片 + 变量 + 数据 + 模型 + 机制 + 识别 + 稳健性 + 异质性 + 表格线索
exp-design
Claim-driven experiment design — scope target claims → design experiment blocks (baseline/validation/ablation/robustness) → build run order → optional Review LLM review → write to wiki
exp-eval
Experiment verdict gate — Review LLM independently judges results → 4 verdict paths → auto-update claims confidence, ideas status, graph edges
ingest
Ingest a paper into the wiki — creates pages (papers + concepts + people + claims) and builds all cross-references and graph edges. Trigger whenever the user says "ingest", "add this paper", drops a `.pdf` / `.tex` / arXiv URL, or asks to fold a paper into the knowledge base.
init
Bootstrap ΩmegaWiki from user sources plus optional discovery, then ingest the final paper set in parallel
novelty
Multi-source novelty verification — WebSearch + Semantic Scholar + wiki + Review LLM cross-verify — outputs novelty score and recommendations
paper-compile
LaTeX compile → PDF — latexmk compile + auto-fix + page count/anonymity/font/[UNCONFIRMED] checks + submission checklist
paper-draft
Draft a LaTeX paper from PAPER_PLAN — write each section from wiki sources + generate figures/tables + BibTeX verification + de-AI polish
paper-plan
Compile a paper outline from the claim graph — evidence map → narrative structure → section plan + figure plan + citation plan, Review LLM review mandatory
prefill
Seed wiki/foundations/ with domain background knowledge so subsequent /ingest does not create duplicate concept pages for textbook material
rebuttal
Parse review comments → atomize concerns (Rvx-Cy) → map to wiki claims → check evidence → Review LLM stress-test → generate rebuttal
refine
General-purpose multi-round iterative improvement — repeatedly calls /review on any research artifact, parses feedback, applies fixes, updates wiki, until the target score is reached
research
端到端研究编排器：idea 发现 → 实验设计 → 执行 → 判决 → 论文撰写，带人工门控和状态恢复
reset
按 scope 重置 wiki 到干净 scaffold（wiki / raw / log / checkpoints / all）。适用于开发迭代或 setup 失败后的无痛重启。
review
通用跨模型审查：Review LLM 对任意研究制品进行独立评审，输出结构化评分、wiki 实体映射与改进建议
survey
从 wiki 知识生成论文 Related Work 章节：主题分组 → 叙事结构 → LaTeX 输出，遵循 citation-verification 和 academic-writing
theory-ingest
将一篇理论建模论文摄取为理论研究 wiki：论文卡片 + 假设/原语 + 命题/定理 + 解概念 + 可检验推论，并与实证层在同一张 graph 上接桥
ask
对 wiki 提问，综合检索相关页面后回答，好的回答可 crystallize 回 wiki
exp-run
实验执行全流程：准备代码 → 部署运行 → 监控状态 → 收集结果，支持三种运行模式
exp-status
查看所有运行中实验的状态，可选自动收集已完成实验并推进流水线
ideate
多阶段研究 idea 生成管道：景观扫描 → 双模型脑暴 → 初筛 → 深度验证 → 写入 wiki
stata-plan
将实证研究设计转成 Stata 执行计划或 do 文件骨架，包含数据合并、变量构造、描述统计、回归和稳健性
check
Scan the full wiki to detect health issues and produce a tiered fix-recommendation report (covers all 9 entity types + graph consistency)
setup
交互式 API key 配置引导 — 检测当前 .env 状态，逐步引导配置 Semantic Scholar、DeepXiv 和 Review LLM
empirical-design
基于已摄取文献和本地数据，生成经管实证研究设计：问题、机制、变量、模型、识别、机制、异质性、稳健性
variable-map
汇总某个经管实证变量在文献中的测算口径、数据来源、模型角色和项目可用性
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.