3. Data, Science & AI

Generates AI-assisted frontmatter field suggestions for Obsidian notes using a structured LLM pipeline for metadata extraction.

Provides step-by-step guidance for creating specialized AI agents (subagents) and defining agent configurations and architecture.

Generates real-time system operation reports by aggregating Cloudflare Workers logs, database queries, and KV cache status.

Generates images and text from prompts via Google Gemini API, serving as a backend for image generation in other skills.

Guides PostgreSQL database management using Prisma for migrations, seeding, and schema evolution.

Designs optimized PostgreSQL schemas using best practices, data types, indexing, and constraints for high-performance database structures.

Manages parallel execution and coordination of multiple AI subagents for complex task delegation and multi-agent workflows.

Architects and implements production-grade tool-calling systems for LLM agents, enabling chatbots to perform actions via external APIs and integrations.

Optimizes Postgres performance and best practices for SQL, migrations, indexes, and RLS policies in Supabase workflows.

Generates reports on Sytex data including tasks, projects, AI credits, and organizational statistics for specified instances.

Creates charts and graphs from CSV, JSON, or query results for data reporting and analysis.

Applies Supabase's best practices to optimize Postgres query performance, schema design, and database configurations.

Transforms data between formats and schemas with explicit loss tracking and validation, ensuring data integrity during conversion.

Discovers hidden patterns, anomalies, and insights in data through exploratory analysis and statistical methods.

Automates merging of multi-source data outputs, resolving conflicts and reconciling constraints to produce unified results for agent coordination or system integration.

Analyzes cause-effect relationships to identify root causes, map dependencies, and build causal models for data-driven decision making.

Quantifies values while providing uncertainty bounds for metrics, risk scores, and measurable properties in data analysis.

Builds rigorous world models incorporating state, dynamics, uncertainty, and provenance for digital twins and simulation foundations.

Detects patterns, entities, or conditions in data for validation, analysis, or signal identification. A fundamental data processing capability.

Recalls past interactions, decisions, and patterns to maintain contextual consistency in AI agent reasoning and decision-making.

Generates domain-specific state representations for modeling system states, entity relationships, and baseline snapshots in AI and data applications.

Ingests and structures external inputs like API responses, webhooks, and sensor data into usable formats for data processing workflows.

Integrates heterogeneous data sources into a unified model with conflict resolution, schema alignment, and provenance tracking for comprehensive data consolidation.

Assigns structured labels to data items based on taxonomy rules for consistent categorization and analysis.