Snowflake-Labs/sfguides
Jupyter Notebook
Captured source
source ↗Snowflake-Labs/sfguides
Description: Tutorials for Snowflake Cortex Code
Language: Jupyter Notebook
License: Apache-2.0
Stars: 5
Forks: 7
Open issues: 1
Created: 2026-02-19T16:51:37Z
Pushed: 2026-03-25T18:30:01Z
Default branch: main
Fork: no
Archived: no
README:
Snowflake Tutorial Skills for Cortex Code
Interactive, agent-guided tutorials that teach Snowflake features through hands-on learning. Built using the "Agent-as-the-Teacher" pattern for Cortex Code.
Quick Install
# Install all tutorial skills with one command npx skills add https://github.com/Snowflake-Labs/sfguides
Or install individual skills:
npx skills add https://github.com/Snowflake-Labs/sfguides/tree/main/dynamic-tables-tutorial npx skills add https://github.com/Snowflake-Labs/sfguides/tree/main/cortex-classify-tutorial npx skills add https://github.com/Snowflake-Labs/sfguides/tree/main/cortex-classify-notebook
> Powered by skills from Vercel Labs
Available Skills
| Skill | Description | |-------|-------------| | dynamic-tables-tutorial | Build data pipelines with automatic refresh, incremental processing, and CDC patterns | | cortex-classify-tutorial | Classify unstructured text using Cortex AI (Python & SQL) | | cortex-classify-notebook | Deploy the CLASSIFY_TEXT tutorial notebook to your Snowflake account |
Usage
Once installed, start a new Cortex Code session and invoke the skill:
You: teach me dynamic tables Agent: [Fetches latest docs, welcomes you, and begins the tutorial]
The agent will: 1. Fetch the latest documentation from Snowflake 2. Detect your environment (prefers SNOWFLAKE_LEARNING if available) 3. Guide you step-by-step through the tutorial 4. Explain before executing any command 5. Pause for confirmation before each step 6. Answer questions using deep reference materials 7. Verify your work at the end
Requirements
- Cortex Code CLI installed
- Snowflake account with Cortex AI enabled
- For best experience: SNOWFLAKE_LEARNING environment (BCR-1992)
Skill Structure
Each skill follows the standard Cortex Code skill format:
skill-name/ ├── SKILL.md # Main agent instructions (YAML frontmatter + markdown) └── references/ # Deep reference materials for answering questions ├── LESSONS.md # Step-by-step code for the tutorial ├── FAQ.md # Quick answers ├── TROUBLESHOOTING.md └── ... # Topic-specific deep dives
Teaching Philosophy
These skills implement the "Agent-as-the-Teacher" pattern:
1. Explain-Before-Execute: Every command is explained before running 2. Pause for Confirmation: User confirms each step (even if auto-allowed) 3. One Step at a Time: Small, digestible chunks 4. Deep Reference Materials: Agent can answer any question thoroughly 5. Verify Understanding: Checkpoint questions between lessons 6. Always Current: Fetches latest docs at tutorial start
Contributing
Contributions welcome! Please ensure:
- SKILL.md follows the Cortex Code skill format
- Include comprehensive reference materials
- Follow the explain-before-execute pattern
- Test with a fresh Snowflake account
License
Apache 2.0 License - see [LICENSE](LICENSE)
Resources
Notability
notability 2.0/10Low stars, routine documentation repo