Pesquisa e Web

Systematically evaluate scholarly work using the ScholarEval framework, providing structured assessment across research quality dimensions including problem formulation, methodology, analysis, and writing with quantitative scoring and actionable feedback.

个人知识库管理工具。支持三个命令：/wiki compile（将 raw/ 素材增量编译为结构化 wiki）、/wiki check（知识审计：有问题列表时做完整审计，无问题列表时做纯结构审计）、/wiki check <questions_file>（基于 wiki 回答问题并做结构审计，合并输出为单份 check 报告）。

Ativa quando o usuário fornece feedback de revisor ou coautor (PDF anotado, comentários colados ou relatório de revisor) e deseja implementar revisões. Extrai pontos de revisão, mapeia-os para locais em paper.tex, classifica ações, implementa mudanças, recompila e gera um registro de alterações + latexdiff, gerenciando o ciclo completo de revisão do feedback às alterações confirmadas.

Access comprehensive LaTeX templates, formatting requirements, and submission guidelines for major scientific publication venues (Nature, Science, PLOS, IEEE, ACM), academic conferences (NeurIPS, ICML, CVPR, CHI), research posters, and grant proposals (NSF, NIH, DOE, DARPA). This skill should be used when preparing manuscripts for journal submission, conference papers, research posters, or grant p

Guided statistical analysis with test selection and reporting. Use when you need help choosing appropriate tests for your data, assumption checking, power analysis, and APA-formatted results. Best for academic research reporting, test selection guidance. For implementing specific models programmatically use statsmodels.

Activate when the user needs to analyze how their paper positions itself relative to the closest existing work. Builds a differentiation matrix, identifies unique positioning, and generates a positioning statement that strengthens the contribution argumentation. Complements the theory-engine by focusing on the competitive landscape of related papers.

Activate when the user needs to evaluate whether a research idea is worth pursuing, brainstorm new research directions, or stress-test a paper concept before committing. This is Phase 0 of the paper machine — the gate that decides whether the full 8-phase pipeline should run. Also activates standalone via /evaluate-idea. Integrates Carlini's research philosophy (conclusion-first test, taste for pr

Perform comprehensive exploratory data analysis on scientific data files across 200+ file formats. This skill should be used when analyzing any scientific data file to understand its structure, content, quality, and characteristics. Automatically detects file type and generates detailed markdown reports with format-specific analysis, quality metrics, and downstream analysis recommendations. Covers

Three AI models, one synthesis — Claude, Gemini & Codex cross-verify each other for rigorous multi-perspective research

Three AI models, one synthesis — Claude, Gemini & Codex cross-verify each other for rigorous multi-perspective research

Autonomous LLM training optimization with GPU support. Runs 5-minute training experiments, measures val_bpb, keeps improvements or reverts — repeat forever. Use this skill when the user asks to "train a model autonomously", "optimize LLM training", "run ML experiments", "autoresearch with GPU", "optimize val_bpb", "autonomous ML training", "LLM pretraining loop", "setup ML autoresearch", "GPU trai

Synthesize a folder of mixed local files (articles, notes, excerpts) into a structured thematic report with key findings and gaps. Use when consolidating research before writing a report, PRD, or strategy doc. Triggers: 'synthesize my research folder', 'research synthesis', 'turn my notes into a report', 'синтезируй папку с исследованиями', 'синтез ресёрча', 'сведи файлы в отчёт'.