Skill UI

This powerful compiler ingests heterogeneous research inputs, including PDF papers, code repositories, experiment logs, and raw notes. It transforms this diverse data into a comprehensive, structured Agent-Native Research Artifact (ARA). The system extracts core knowledge atoms, builds cognitive maps (claims, concepts, heuristics), and generates a verifiable, executable knowledge package for advanced AI consumption and research iteration.

Expert assistance for building, modifying, and debugging Minecraft server plugins for the Paper, Spigot, and Bukkit ecosystems. This includes implementing complex gameplay mechanics such as minigames, advanced combat systems, persistent player progression (perks, economy, quests), custom command handling, and robust state management. We handle configuration, listener registration, and adherence to best practices for Java-based server API usage.

Generates a low-fidelity, hand-drawn wireframe sketch perfect for communicating early design concepts. It meticulously mimics the look of marker drawings, sticky notes, and whiteboard scribbles on graph paper, ensuring the output reads like a preliminary ideation sketch rather than a finished UI. Ideal for use when the brief specifies 'lo-fi,' 'sketch,' or 'whiteboard.'

A comprehensive tool for searching academic literature using the OpenAlex API. Retrieve detailed metadata, including open citation graphs, author institutional affiliations, and funding sources. Ideal for deep, systematic academic research when basic preprint or citation data is insufficient.

This skill simulates a hostile, two-step review process to stress-test a paper's core claims. First, an adversarial reviewer constructs the single strongest rejection memo. Second, a fresh reviewer defends the paper point-by-point, identifying critical, unresolved weaknesses that standard reviews often miss. Ideal for pre-submission validation or rebuttal preparation.

This skill automates the complex process of resubmitting an already polished academic paper to a different conference or journal. It is designed for scenarios where the core content, experiments, and bibliography must remain untouched, but the paper needs micro-edits to accommodate new venue constraints, page limits, or prior reviewer feedback. It ensures the new submission is generated as a separate, sibling directory to maintain the integrity of prior submissions.

A comprehensive server providing research-grade access to arXiv. Users can search papers by topic, author, or date, download full texts, and analyze the scientific landscape by traversing citation graphs. It also supports local semantic search over downloaded documents and enables monitoring for new papers with real-time alerts. Ideal for academic research and literature review.

This end-to-end pipeline automates the conversion of a completed academic research paper into a full conference talk artifact. It generates a slide outline, Beamer source, editable PPTX, speaker notes, full script, and polished visuals. Critically, it includes rigorous assurance checks (citation, claim, anonymity) to ensure academic integrity, making the output suitable for top-tier conferences like NeurIPS or ICML. Use this when you need the complete, polished presentation package, not just a draft deck.

This skill acts as an expert guide for authoring comprehensive, implementation-ready design system documentation. It structures guidelines covering foundational design tokens (color, spacing, typography), detailed component anatomy, interaction states, and WCAG accessibility standards. Ideal for product designers and engineers needing to define a cohesive and scalable UI language.

This comprehensive 12-agent pipeline guides users through the entire academic writing lifecycle, from initial research planning to final publication-ready draft. Features include literature review, structure design, full-text drafting, citation compliance (multiple formats), bilingual abstract generation, and simulated peer review. It ensures high writing quality via style calibration and advanced checks, supporting output formats like LaTeX, DOCX, and PDF.

Simulates a comprehensive, multi-stage academic journal peer review process. This agent dynamically configures five expert reviewers (EIC, 3 peers, and a Devil's Advocate) to evaluate a paper across methodology, domain expertise, cross-disciplinary impact, and core arguments. It provides structured reports, editorial decisions, and revision roadmaps, supporting full reviews, post-revision verification, and calibration modes.

This orchestrator manages the entire academic research lifecycle, guiding users from initial research exploration through writing, multi-stage peer review, rigorous integrity checks, and final publication preparation. It coordinates specialized skills (e.g., deep-research, paper generation, reviewer feedback integration) across a mandatory 10-stage workflow, ensuring reproducibility and academic rigor at every step. It also supports resuming complex workflows from previously saved checkpoints (Material Passports).