VS

vinodhini-sd/build-with-coco

Developer tools
23 stars Quality 37 Trend 31

A collection of reusable Cortex Code skills and copy-paste prompts for automating data engineering workflows on Snowflake.

Overview

A collection of reusable Cortex Code skills and copy-paste prompts for automating data engineering workflows on Snowflake. Copy-paste prompts in two categories — browse by role or by scenario. Full catalog in recipes/README.md. (recipes/by-role/): data-engineer, analytics-engineer, data-analyst, ai-ml-engineer, app-developer, data-governance-lead, snowflake-admin Skills are markdown-based workflow definitions that teach Cortex Code how to perform complex, multi-step tasks. They provide structured guidance, tool references, decision logic, and user checkpoints. Copy a skill or prompt directory into your global folder: Once installed, invoke a skill in Cortex Code with: Or simply describe what you need — Cortex Code will match your request to the right skill based on trigger phrases. For prompts, just paste them directly into Cortex Code CLI. See each prompt file for the copy-paste block and customization tips.

README

build-with-coco

A collection of reusable Cortex Code skills and copy-paste prompts for automating data engineering workflows on Snowflake.

Skills

Skill Description Trigger Phrases
dbt-model-generator Auto-generate dbt dimensional models (star schema) from raw Snowflake tables “generate dbt models”, “shift left”, “dimensional model”, “star schema from raw”
developer-voice Research real developer sentiment from Reddit and community forums, then synthesize into talk tracks, research briefs, or competitive intel grounded in practitioner quotes “developer voice”, “community pulse”, “reddit research”, “talk track”, “what are developers saying”, “pain points”
know-your-data Discover data you already have access to in a Snowflake account, understand what it contains, and map it to your roles “know your data”, “know my data”, “find data”, “data discovery”, “what can I access”, “explore account”
poc-builder Go from zero to a working POC with any Snowflake guide, blog post, doc link, or topic name. Deep-dives the content, extracts a demo spec, finds matching data in your account, and builds a working POC interactively “poc builder”, “build this”, “zero to poc”, “try this”, “teach me”, “walk me through”, “prototype this”, “run this guide”
aws-glue-iceberg-setup Set up AWS Glue as an Iceberg catalog for Snowflake — crawl S3 data, register tables, configure external volumes, and query via Iceberg “glue iceberg”, “aws glue setup”, “iceberg from s3”, “external catalog”, “glue crawler”
data-quality-suite Multi-framework DQ suite — set up quality checks, monitoring, and pipeline gates across Snowflake DMFs, dbt, Great Expectations, Soda Core, Snowpark, and PySpark. Fork customer-config.md to customize for your team “data quality”, “quality checks”, “quality gates”, “dbt tests”, “soda checks”, “great expectations”, “dq suite”

Recipes

Copy-paste prompts in two categories — browse by role or by scenario. Full catalog in recipes/README.md.

By role (recipes/by-role/): data-engineer, analytics-engineer, data-analyst, ai-ml-engineer, app-developer, data-governance-lead, snowflake-admin

Repo Structure

build-with-coco/
├── skills/
│   ├── dbt-model-generator/
│   │   ├── SKILL.md
│   │   ├── COMPASS.md           # 25–35 line navigation guide
│   │   └── references/
│   │       └── workflow.md
│   ├── developer-voice/
│   │   ├── SKILL.md
│   │   ├── COMPASS.md
│   │   └── references/
│   │       └── html-styling.md
│   ├── know-your-data/
│   │   ├── SKILL.md
│   │   └── COMPASS.md
│   ├── aws-glue-iceberg-setup/
│   │   ├── SKILL.md
│   │   ├── COMPASS.md
│   │   └── references/
│   └── poc-builder/
│       ├── SKILL.md
│       ├── COMPASS.md
│       └── references/
│           ├── ACCOUNT_DISCOVERY.md
│           ├── BUILD_SUMMARY_TEMPLATE.md
│           ├── GUIDE_PARSING.md
│           ├── KNOWN_GOTCHAS.md
│           └── TEACHING_PATTERNS.md
├── recipes/
│   ├── by-role/             # prompts organized by job role
│   ├── connect-sources/
│   ├── build-pipelines/
│   ├── monitor-quality/
│   ├── optimize-costs/
│   ├── secure-and-govern/
│   ├── assess-change-impact/
│   ├── self-serve-analytics/
│   ├── ai-enrichment/
│   └── README.md
├── docs/
│   └── COMPASS_GUIDE.md     # how to write COMPASS.md files
├── AGENTS.md                # agent contribution guide
├── install.sh
└── hooks/                   # session hooks (optional)
    ├── check-errors.py
    ├── session-start.sh
    ├── set-tab-title.sh
    ├── tab-title-helper.py
    ├── whats-new-helper.py
    └── README.md

What Are Cortex Code Skills?

Skills are markdown-based workflow definitions that teach Cortex Code how to perform complex, multi-step tasks. They provide structured guidance, tool references, decision logic, and user checkpoints.

Installation

Quick install (everything)

./install.sh

Skills only or recipes only

./install.sh skills    # skills only
./install.sh recipes   # recipes only

Per-project install

./install.sh --project   # installs into .cortex/ in current directory

Manual

Copy a skill or prompt directory into your global folder:

cp -r skills/dbt-model-generator ~/.snowflake/cortex/skills/dbt-model-generator

Usage

Once installed, invoke a skill in Cortex Code with:

$dbt-model-generator

Or simply describe what you need — Cortex Code will match your request to the right skill based on trigger phrases.

For prompts, just paste them directly into Cortex Code CLI. See each prompt file for the copy-paste block and customization tips.

Contributing

To add a new skill:

  1. Create a directory under skills//
  2. Add a SKILL.md following the Cortex Code skill format
  3. If SKILL.md exceeds 150 lines, add a COMPASS.md (see docs/COMPASS_GUIDE.md)
  4. Update this README with the skill entry
  5. Submit a PR
View this README on GitHub

Recommended Tools

Try a different keyword or remove a filter.

Install

npx skillfish add vinodhini-sd/build-with-coco