jamescha-earley

Guide to Snowflake Cortex AI functions — SQL-native LLM inference, text classification, extraction, sentiment analysis, embeddings, and document processing without infrastructure management

Best practices for Snowflake Dynamic Tables — declarative data pipelines with automatic incremental refresh, replacing complex Streams + Tasks patterns

MUST USE when reviewing Snowflake schemas, queries, warehouses, or data pipelines. Contains 25 rules that MUST be checked before providing recommendations. Always read relevant rule files and cite specific rules in responses. Triggers: snowflake, create table, warehouse, clustering key, query optimization, data loading, copy into, variant, semi-structured, cost optimization, snowflake security.

Guide to Snowflake Postgres — managed PostgreSQL with pg_lake (Iceberg tables on object storage), pg_incremental (automated archival), and zero-ETL bridge to Snowflake analytics

Best practices for building Snowpark Python applications — DataFrame API, UDFs, UDTFs, stored procedures, and deployment patterns

Containerize an application with a production-grade Dockerfile, Docker Compose for local development, and optional Dev Container configuration. Use for: Dockerizing apps, multi-stage builds, Compose local dev stacks, .dockerignore, image size optimization, dev containers. Triggers: docker, dockerfile, compose, container, containerize, dockerize, docker-compose, devcontainer, dev container, docker setup, docker image, docker build.

Scaffold a Drizzle ORM project with TypeScript schema, relations, database client, and migrations. Use for: setting up Drizzle from scratch, designing table schemas, configuring drizzle-kit, writing type-safe queries, multi-database support (PostgreSQL, MySQL, SQLite). Triggers: drizzle, drizzle orm, drizzle setup, drizzle schema, drizzle migration, drizzle-kit, typescript orm, sql-like orm, drizzle config, drizzle relations, drizzle project.

Scaffold a Supabase project with database schema, Row Level Security policies, and auth integration. Use for: setting up RLS policies, debugging 'RLS not working', multi-tenant access control, auth.uid() patterns, team/org membership, profile triggers. Triggers: supabase, supabase rls, row level security, supabase auth, rls policy, supabase setup, supabase permissions, multi-tenant supabase, supabase project, rls not working.

Set up the good-morning daily briefing skill from scratch. Walks through Google Workspace MCP installation, Snowflake storage setup, and Google Doc tracking configuration. Use when: user wants to set up good morning, create a daily briefing, configure morning briefing, install good morning skill. Triggers: setup good morning, install good morning, create briefing, setup daily briefing, configure morning briefing, setup-good-morning.

Build automated Cortex Code demo recordings end-to-end. Walks through an interactive wizard to plan a demo scenario, synthesizes realistic sample data (tables, documents, stages) in Snowflake, generates a YAML prompts file, then launches a real cortex session in tmux that types each prompt with human-like speed, auto-accepts permissions, waits for responses, and records the terminal as an MP4 via ffmpeg screen capture. Produces three artifacts: cortex_demo_prompts.yaml (the demo script), cortex_demo_setup.sql (idempotent Snowflake setup), and cortex_demo_setup.py (data synthesis). Use when: user wants to create a demo, build a demo, generate demo prompts, set up a demo environment, prepare a demo, create sample data for a demo, synthesize demo data, build a showcase, create a walkthrough, generate a cortex demo, or record a demo video. Triggers: demo builder, build demo, create demo, demo prompts, demo yaml, demo setup, synthesize data, demo data, showcase, walkthrough, cortex demo, record demo, demo video, d

Build automated Cortex Code demo recordings end-to-end. Walks through an interactive wizard to plan a demo scenario, synthesizes realistic sample data (tables, documents, stages) in Snowflake, generates a YAML prompts file, then launches a real cortex session in tmux that types each prompt with human-like speed, auto-accepts permissions, waits for responses, and records the terminal as an MP4 via ffmpeg screen capture. Produces three artifacts: cortex_demo_prompts.yaml (the demo script), cortex_demo_setup.sql (idempotent Snowflake setup), and cortex_demo_setup.py (data synthesis). Use when: user wants to create a demo, build a demo, generate demo prompts, set up a demo environment, prepare a demo, create sample data for a demo, synthesize demo data, build a showcase, create a walkthrough, generate a cortex demo, or record a demo video. Triggers: demo builder, build demo, create demo, demo prompts, demo yaml, demo setup, synthesize data, demo data, showcase, walkthrough, cortex demo, record demo, demo video, d

Switch the active Snowflake agent connection. Use when: user wants to change which Snowflake account the Cortex Code agent uses, switch the agent's connection, or point the agent at a different environment. Does NOT handle SQL-level connection objects (CREATE CONNECTION) or general account switching. Triggers: change agent connection, switch agent connection, cortex agent connection, cortexAgentConnectionName, agent connection name.

Rewrite text to remove AI slop, tropes, and cliches. Makes text more genuine, natural, and human-sounding. Use when: user asks to deslopify, naturalize, de-slop, remove AI tropes, make text sound human, clean up AI writing, or review text for AI patterns. Triggers: deslopify, naturalize, de-slop, AI slop, AI tropes, sound human, AI writing.