Skill UI

DeepChem is a Python library for molecular machine learning, providing featurizers, dataset loaders, splitters, and models spanning traditional ML to GNNs for property prediction, drug discovery, and materials design.

This meta-skill is designed to orchestrate the creation of complex, multi-step workflows (meta-skills). It should only be used when a user explicitly requests combining, synthesizing, or proposing a new skill that coordinates multiple existing functions. The process includes intent clarification, trigger collision checking, validation, and generating a detailed proposal containing the workflow DAG.

A detailed guide for authors submitting theoretical machine learning papers (like COLT) regarding the appropriate inclusion and design of numerical simulations. It emphasizes that numerics should only visualize proved bounds, phase transitions, or separations, maintaining the focus on theoretical proofs rather than empirical data. Provides rules for plotting, reporting variability, and maintaining academic honesty.

This guide outlines the rigorous methodological standards for empirical work in AI fairness, accountability, and transparency (FAccT). It advises researchers on matching specific claims (e.g., disparate harm) to appropriate evidence types—including disaggregated metrics, coded qualitative data, and documented ethical procedures. It stresses the importance of ethical review, handling protected attributes, and acknowledging limitations to ensure high scientific and ethical rigor.

This comprehensive audit checklist is designed for authors submitting papers to the ACM RecSys conference. It guides users through selecting the correct specialized track (Main, Industry, Reproducibility), adhering to strict page budgets, ensuring thorough double-blind anonymization, and mitigating common submission risks like identity leaks and improper data splitting.

A systematic post-hoc audit skill for verifying factual, legal, and interpretive claims generated by AI. Designed for professional reliance—such as legal briefs or published advice—it classifies claims by type (empirical, legal, interpretive) and salience, searches for corroborating evidence using tiered, domain-appropriate sources (e.g., NICE guidelines, primary legislation), and scores the trustworthiness of the output before professional use.

This skill focuses on designing tools for AI agents to interact with the real world effectively. It covers schema design, JSON Schema best practices, error handling, and validation. It also addresses description writing for LLMs and the emerging MCP standard. Proper tool design prevents hallucinations and ensures safety. Includes Python SDK examples for function calling.

This skill simulates the persona of Geoffrey Hinton, known as the Godfather of Deep Learning. It is designed for discussions on neural networks, backpropagation, and deep belief networks. Users can engage in expert-level conversations about AI history and theory. Note that this is a fictional perspective exercise and not actual communication with Hinton.

Master the art of evaluating AI, LLM, and agent systems. Learn to decide when an evaluation is needed, design robust experiments, choose appropriate graders, and interpret results under hostile audit. Covers everything from spot checks to CI gates and benchmarking.