Lingzhi Yang builds skills that help AI agents carry out the research pipeline, from searching literature and generating and checking research ideas to writing literature surveys, debugging experiment code, and analyzing experimental data with statistical tests. Yang's skills also cover equation discovery through LLM-guided symbolic regression, revising papers against reviewer feedback, and drawing and refining diagrams on a live Excalidraw canvas.
Agent Research Skills
31 skills · 451 stars
Algorithm Design Design algorithms with LaTeX pseudocode and UML diagrams.
Atomic Decomposition Decompose research ideas into atomic, self-contained concepts with bidirectional math-code mapping.
Backward Traceability Make every number in the final PDF traceable to the exact code line that produced it.
Citation Management Manage BibTeX citations for LaTeX papers.
Code Debugging Debug experiment code with structured error analysis.
Data Analysis Generate statistical analysis code with 4-round review.
Deep Research Conduct systematic academic literature reviews in 6 phases, producing structured notes, a curated…
Excalidraw Skill Programmatic canvas toolkit for creating, editing, and refining Excalidraw diagrams via MCP tools…
Experiment Code Write ML experiment code with iterative improvement.
Experiment Design Design experiment plans with progressive stages — initial implementation, baseline tuning, creati…
Figure Generation Generate publication-quality scientific figures using matplotlib/seaborn with a three-phase pipel…
Github Research Explore and analyze GitHub repositories related to a research topic.
Idea Generation Generate novel research ideas with iterative refinement and novelty checking against literature.
Latex Formatting Handle LaTeX formatting, templates, and styling for academic papers.
Literature Review Conduct comprehensive literature reviews using multi-perspective dialogue simulation.
Literature Search Search academic literature using Semantic Scholar, arXiv, and OpenAlex APIs.
Math Reasoning Formal mathematical reasoning for research papers — derive equations, write proofs, formalize pro…
Novelty Assessment Assess research idea novelty through systematic literature search.
Paper Assembly Orchestrate the full paper pipeline end-to-end.
Paper Compilation Compile LaTeX papers to PDF with automatic error detection, chktex style checking, and citation/r…
Paper Revision Revise papers based on reviewer feedback.
Paper To Code Convert an ML research paper into a complete, runnable code repository.
Paper Writing Section Write a specific section of an academic paper (Abstract, Introduction, Background, Related Work,…
Rebuttal Writing Write point-by-point rebuttals to reviewer comments.
Related Work Writing Write Related Work sections that compare and contrast prior work with your approach.
Research Planning Design research plans and paper architectures.
Self Review Automatically review an academic paper using the NeurIPS review form with three reviewer personas…
Slide Generation Convert a completed paper into presentation slides (Beamer LaTeX) or poster.
Survey Generation Generate complete academic survey papers using multi-LLM parallel outline generation, RAG-based s…
Symbolic Equation Discover scientific equations from data using LLM-guided evolutionary search (LLM-SR).
Table Generation Generate publication-quality LaTeX tables from experimental results.