Skill UI

This comprehensive skill bridges the gap between initial idea discovery and final model deployment. It ingests a structured experiment plan (EXPERIMENT_PLAN.md), automatically implements the required code, performs cross-model code review, deploys the setup to GPU, and collects initial quantitative results, allowing for a rapid iteration loop in AI research.

A comprehensive guide for designing, reviewing, and migrating numerical workflows in MATLAB or GNU Octave. It emphasizes building reproducible projects, adhering to strict safety boundaries, and ensuring code semantics are explicit for robust data processing and analysis.

This guide identifies and provides solutions for common anti-patterns and mistakes when working with Snowflake. It covers best practices in SQL writing, warehouse resource management, data loading strategies (e.g., using MERGE), and access control setup. It is essential for auditing existing deployments, onboarding new users, or performing code reviews to ensure cost efficiency and data integrity.

A comprehensive guide for structuring research artifacts (code, models, datasets, media) for ACM Multimedia conferences. It details the distinct requirements for anonymous review packages versus permanent public releases, offering guidelines on ethics, licensing, reproducibility, and track-specific documentation.

A comprehensive guide for assembling a fully reproducible research package for macroeconomics papers targeting the American Economic Journal (AEJ: Macro). It ensures compliance with the AEA Data and Code Availability Policy by documenting raw data, simulation/calibration code, estimation routines, and visualization steps, guaranteeing that results can be independently verified by reviewers.

This guide outlines the comprehensive package structure required for submitting manuscripts to American Economic Review: Insights (AER: Insights). It ensures authors meet the AEA Data and Code Availability Policy, providing all necessary code, raw data access, and documentation (including a master script and exhibit-to-code map) to allow the data editor to verify the reproducibility of every result before publication. It is crucial for passing the final publication gate.

This guide provides comprehensive best practices for preparing supplementary materials for major academic conferences like AISTATS. It details the proper structure and handling of appendices, mathematical proofs, code archives, and additional data tables, ensuring that the main paper remains fully understandable even if the supplementary material is ignored by reviewers. Essential for publishing high-quality, robust research.

A comprehensive guide for strengthening the methodological rigor of academic manuscripts, particularly for high-stakes journals. It covers advanced techniques for causal identification (e.g., Difference-in-Differences, Instrumental Variables, Regression Discontinuity), pre-analysis planning, linking theory to empirical tests, and defending claims against skeptical reviewers. This skill focuses on methodological architecture, not code.

This guide outlines a systematic methodology for conducting comprehensive literature reviews in economics. It moves beyond simple summarizing by teaching researchers how to build an evidence matrix, ensuring systematic coverage by sweeping databases (JEL codes, keywords, working papers). It is essential for preparing high-impact academic reviews that must prove exhaustive coverage and synthesize complex, often conflicting, findings.

A methodological guide for documenting the rigor of literature reviews, especially in sociology. It details how to build a comprehensive 'coverage account' (a near-PRISMA record of the search process) and how to ensure any quantitative synthesis (meta-analysis, bibliometrics) conducted within the review is fully reproducible by depositing data and code.

A comprehensive guide to adhering to ASA/ASR data-sharing and transparency standards for social science research. It helps users prepare thorough documentation, define data provenance (codebooks, scripts), manage confidentiality constraints (IRB, proprietary data), and ethically prepare materials for academic publication. Use this when structuring a manuscript's data section or anticipating reviewer questions about data verification.

This guide outlines best practices for packaging research artifacts (prototypes, datasets, instruments, code) for academic review and permanent public release, particularly in HCI venues like CHI. It covers strategies for creating highly reviewable, anonymous submissions (e.g., claims maps, minimal demos) and establishing robust, licensed, permanent archives (e.g., DOIs, OSF), ensuring research reproducibility and auditability.