3. Data, Science & AI

Automates web interactions using Node.js and LLM-driven element selection, with deterministic playback, cost controls, and fallback paths to replace expensive UI exploration tools.

Manages exploratory computational analysis sessions for data science, scientific computing, and machine learning experiments, tracking iterative results.

Provides universal context management for AI agents with proactive compaction and summarization to optimize processing efficiency.

Formats dates/times, calculates differences, and provides day-of-week information for date-related queries.

Guides AI agent behavior using HHH framework (Helpful, Honest, Harmless) with exploration-before-implementation and anti-sycophancy rules for all interactions.

Performs all mathematical operations including arithmetic, percentages, averages, and date/calendar calculations for numeric analysis.

View and track personal health metrics including activity, sleep, heart rate, and wellness data via Fitbit integration.

Provides patterns for hierarchical Bayesian models including random effects, partial pooling, and parameterization techniques.

Automates user behavior tracking and analytics using Heap's auto-capture platform.

Enables Bayesian meta-analysis with fixed, random, and network models using Stan and JAGS for statistical synthesis of research data.

Core functions for time-to-event clinical trial simulation, including survival data generation and weighted logrank tests.

Enables event-based tracking and analysis of user behavior and product metrics using Mixpanel's analytics platform.

Manages MySQL database operations including schema, migrations, queries, and backup/restore for The Dreamer's Cave application.

Command-line tool for semantic vector search on Agentuity cloud platform, requiring authentication for secure data retrieval operations.

Provides PostgreSQL patterns for Neon serverless database configuration, query writing, and optimization for Todo backend applications.

Provides deep methodology knowledge for MAIC analyses, including assumptions, weight diagnostics, and ESS interpretation for causal inference studies.

Provides foundational knowledge for writing Bayesian statistical models using Stan 2.37, covering structure, types, distributions, and best practices.

Analyzes product usage and user behavior data using Amplitude's platform for actionable business insights.

AI meeting assistant that transcribes audio and analyzes content for actionable insights and summaries.

Provides tradeoff-based optimization strategies for clinical trial design, including sample size calculation and sensitivity analysis to maximize statistical power.

Provides Bayesian time series modeling capabilities using AR, MA, ARMA, state-space, and dynamic linear models in Stan and JAGS.

Foundational knowledge for writing BUGS/JAGS models: declarative syntax, precision parameterization, distributions, and R integration.

Performs MCMC diagnostics for Bayesian models, assessing convergence, effective sample size, divergences, and posterior predictive checks.

Provides foundational knowledge for writing PyMC 5 models, including syntax, distributions, sampling, and ArviZ diagnostics.