@lingzhi227

data-analysis provides a structured workflow for generating and reviewing statistical analysis code across four verification rounds, with built-in test selection logic and p-value formatting utilities.

backward-traceability embeds hyperlinked anchors in pdf output so every numeric result connects to its source code line, enabling readers to verify calculations and origins through clickable references.

paper-to-code transforms ml research papers into executable repositories through a four-stage pipeline (planning, analysis, coding, debugging) that extracts methodology, designs architecture via uml diagrams, and generates fully-implemented dependency-ordered code.

citation-management handles the full lifecycle of latex bibliography maintenance: harvesting missing citations from semantic scholar, validating cite keys and bib entries, deduplicating, and formatting bibtex. covers four distinct actions (harvest, validate, add, format) with concrete command examples and key rules.

self-review applies three independent reviewer personas to academic papers, aggregates scores with weighted averaging, and produces a structured report with consensus findings and actionable feedback.

github-research discovers and analyzes open-source repositories related to a research topic by consuming deep-research outputs and producing integration guidance, though lacks detailed step-by-step procedures and error handling coverage.

experiment-code generates and iteratively improves ml experiment pipelines with structured output directories, logging, and visualization. covers code generation, debugging, and result plotting via four distinct actions.

math-reasoning guides symbolic and formal mathematical work: equation derivations with justified steps, theorem proofs across multiple techniques, problem formalization with explicit variable domains, statistical test selection, and latex generation for publication readiness.

literature-search queries semantic scholar, arxiv, and openAlex for academic papers, outputting jsonl with metadata. incomplete implementation with partial script exposure and no failure handling.

experiment-design structures research validation into four progressive stages: baseline, tuning, creative research, and ablation. accepts research ideas and outputs staged experiment plans with dataset and metric guidance.

symbolic-equation applies llm-guided evolutionary search to discover interpretable scientific equations from data, but the skill document remains incomplete with vague workflow and missing decision logic.

Conduct comprehensive literature reviews using multi-perspective dialogue simulation. Generate diverse expert personas, conduct grounded Q&A conversations, and…

Generate complete academic survey papers using multi-LLM parallel outline generation, RAG-based subsection writing, citation validation, and local coherence…

Assess research idea novelty through systematic literature search. Multi-round search-evaluate loops with harsh critic persona. Binary novel/not-novel decision…

Programmatic canvas toolkit for creating, editing, and refining Excalidraw diagrams via MCP tools with real-time canvas sync. Use when an agent needs to (1)…

Write point-by-point rebuttals to reviewer comments. Extract concerns from reviews, generate evidence-based responses, and format as a structured rebuttal…

Write Related Work sections that compare and contrast prior work with your approach. Organize by theme, cite broadly, and explain how your work differs. Use…

Design algorithms with LaTeX pseudocode and UML diagrams. Generate algorithmic environments, Mermaid class/sequence diagrams, and ensure consistency between…

Revise papers based on reviewer feedback. Map reviewer concerns to specific sections, apply targeted edits, run additional experiments if needed, and verify…

Generate novel research ideas with iterative refinement and novelty checking against literature. Score ideas on Interestingness, Feasibility, and Novelty. Use…

Write a specific section of an academic paper (Abstract, Introduction, Background, Related Work, Methods, Experiments, Results, Discussion/Conclusion) with…

Compile LaTeX papers to PDF with automatic error detection, chktex style checking, and citation/reference validation. Runs the full pdflatex + bibtex pipeline.…

Generate publication-quality scientific figures using matplotlib/seaborn with a three-phase pipeline (query expansion, code generation with execution, VLM…

Handle LaTeX formatting, templates, and styling for academic papers. Set up conference templates (ICML, ICLR, NeurIPS, AAAI, ACL), fix formatting issues,…

Generate publication-quality LaTeX tables from experimental results. Convert JSON/CSV data to booktabs-styled tables with bold best results, multi-row layouts,…

Orchestrate the full paper pipeline end-to-end. Manage state propagation between phases (literature → plan → code → experiments → figures → tables → writing →…

Design research plans and paper architectures. Given a research topic or idea, generate structured plans with methodology outlines, paper structure,…

Debug experiment code with structured error analysis. Categorize errors, apply targeted fixes with retry logic, and use reflection to prevent recurring issues.…

Convert a completed paper into presentation slides (Beamer LaTeX) or poster. Extract key figures, tables, equations, and create a narrative flow for oral…

atomic-decomposition — an installable skill for AI agents, published by lingzhi227/agent-research-skills.

deep-research — an installable skill for AI agents, published by lingzhi227/agent-research-skills.