Snowflake Platform Skill

Build and deploy applications on Snowflake's AI Data Cloud using the snow CLI, Cortex AI functions, Native Apps, and Snowpark.

Quick Start

Install Snowflake CLI

pip install snowflake-cli
snow --version  # Should show 3.14.0+

Configure Connection

# Interactive setup
snow connection add

# Or create ~/.snowflake/config.toml manually

[connections.default]
account = "orgname-accountname"
user = "USERNAME"
authenticator = "SNOWFLAKE_JWT"
private_key_path = "~/.snowflake/rsa_key.p8"

Test Connection

snow connection test -c default
snow sql -q "SELECT CURRENT_USER(), CURRENT_ACCOUNT()"

When to Use This Skill

Use when:

• Building applications on Snowflake platform
• Using Cortex AI functions in SQL queries
• Developing Native Apps for Marketplace
• Setting up JWT key-pair authentication
• Working with Snowpark Python

Don't use when:

• Building Streamlit apps (use streamlit-snowflake skill)
• Need data engineering/ETL patterns
• Working with BI tools (Tableau, Looker)

Cortex AI Functions

Snowflake Cortex provides LLM capabilities directly in SQL. Functions are in the SNOWFLAKE.CORTEX schema.

Core Functions

Function	Purpose	GA Status
COMPLETE / AI_COMPLETE	Text generation from prompt	GA Nov 2025
SUMMARIZE / AI_SUMMARIZE	Summarize text	GA
TRANSLATE / AI_TRANSLATE	Translate between languages	GA Sep 2025
SENTIMENT / AI_SENTIMENT	Sentiment analysis	GA Jul 2025
AI_FILTER	Natural language filtering	GA Nov 2025
AI_CLASSIFY	Categorize text/images	GA Nov 2025
AI_AGG	Aggregate insights across rows	GA Nov 2025

COMPLETE Function

-- Simple prompt
SELECT SNOWFLAKE.CORTEX.COMPLETE(
    'llama3.1-70b',
    'Explain quantum computing in one sentence'
) AS response;

-- With conversation history
SELECT SNOWFLAKE.CORTEX.COMPLETE(
    'llama3.1-70b',
    [
        {'role': 'system', 'content': 'You are a helpful assistant'},
        {'role': 'user', 'content': 'What is Snowflake?'}
    ]
) AS response;

-- With options
SELECT SNOWFLAKE.CORTEX.COMPLETE(
    'mistral-large2',
    'Summarize this document',
    {'temperature': 0.3, 'max_tokens': 500}
) AS response;

Available Models:

• llama3.1-70b, llama3.1-8b, llama3.2-3b
• mistral-large2, mistral-7b
• snowflake-arctic
• gemma-7b
• claude-3-5-sonnet (200K context)

Model Context Windows (Updated 2025):

Model	Context Window	Best For
Claude 3.5 Sonnet	200,000 tokens	Large documents, long conversations
Llama3.1-70b	128,000 tokens	Complex reasoning, medium documents
Llama3.1-8b	8,000 tokens	Simple tasks, short text
Llama3.2-3b	8,000 tokens	Fast inference, minimal text
Mistral-large2	Variable	Check current docs
Snowflake Arctic	Variable	Check current docs

Token Math: ~4 characters = 1 token. A 32,000 character document ≈ 8,000 tokens.

Error: Input exceeds context window limit → Use smaller model or chunk your input.

SUMMARIZE Function

-- Single text
SELECT SNOWFLAKE.CORTEX.SUMMARIZE(article_text) AS summary
FROM articles
LIMIT 10;

-- Aggregate across rows (no context window limit)
SELECT AI_SUMMARIZE_AGG(review_text) AS all_reviews_summary
FROM product_reviews
WHERE product_id = 123;

TRANSLATE Function

-- Translate to English (auto-detect source)
SELECT SNOWFLAKE.CORTEX.TRANSLATE(
    review_text,
    '',      -- Empty = auto-detect source language
    'en'     -- Target language
) AS translated
FROM international_reviews;

-- Explicit source language
SELECT AI_TRANSLATE(
    description,
    'es',    -- Source: Spanish
    'en'     -- Target: English
) AS translated
FROM spanish_products;

AI_FILTER (Natural Language Filtering)

Performance: As of September 2025, AI_FILTER includes automatic optimization delivering 2-10x speedup and up to 60% token reduction for suitable queries.

-- Filter with plain English
SELECT * FROM customer_feedback
WHERE AI_FILTER(
    feedback_text,
    'mentions shipping problems or delivery delays'
);

-- Combine with SQL predicates for maximum optimization
-- Query planner applies standard filters FIRST, then AI on smaller dataset
SELECT * FROM support_tickets
WHERE created_date > '2025-01-01'  -- Standard filter applied first
  AND AI_FILTER(description, 'customer is angry or frustrated');

Best Practice: Always combine AI_FILTER with traditional SQL predicates (date ranges, categories, etc.) to reduce the dataset before AI processing. This maximizes the automatic optimization benefits.

Throttling: During peak usage, AI function requests may be throttled with retry-able errors. Implement exponential backoff for production applications (see Known Issue #10).

AI_CLASSIFY

-- Categorize support tickets
SELECT
    ticket_id,
    AI_CLASSIFY(
        description,
        ['billing', 'technical', 'shipping', 'other']
    ) AS category
FROM support_tickets;

Billing

Cortex AI functions bill based on tokens:

• ~4 characters = 1 token
• Both input AND output tokens are billed
• Rates vary by model (larger models cost more)

Cost Management at Scale (Community-sourced):

Real-world production case study showed a single AI_COMPLETE query processing 1.18 billion records cost nearly $5K in credits. Cost drivers to watch:

1. Cross-region inference: Models not available in your region incur additional data transfer costs
2. Warehouse idle time: Unused compute still bills, but aggressive auto-suspend adds resume overhead
3. Large table joins: Complex queries with AI functions multiply costs

-- This seemingly simple query can be expensive at scale
SELECT
    product_id,
    AI_COMPLETE('mistral-large2', 'Summarize: ' || review_text) as summary
FROM product_reviews  -- 1 billion rows
WHERE created_date > '2024-01-01';

-- Cost = (input tokens + output tokens) × row count × model rate
-- At scale, this adds up fast

Best Practices:

• Filter datasets BEFORE applying AI functions
• Right-size warehouses (don't over-provision)
• Monitor credit consumption with QUERY_HISTORY views
• Consider batch processing instead of row-by-row AI operations

Source: The Hidden Cost of Snowflake Cortex AI (Community blog with billing evidence)

Authentication

JWT Key-Pair Authentication

Critical: Snowflake uses TWO account identifier formats:

Format	Example	Used For
Organization-Account	irjoewf-wq46213	REST API URLs, connection config
Account Locator	NZ90655	JWT claims (iss, sub)

These are NOT interchangeable!

Discover Your Account Locator

SELECT CURRENT_ACCOUNT();  -- Returns: NZ90655

Generate RSA Key Pair

# Generate private key (PKCS#8 format required)
openssl genrsa 2048 | openssl pkcs8 -topk8 -inform PEM -out ~/.snowflake/rsa_key.p8 -nocrypt

# Generate public key
openssl rsa -in ~/.snowflake/rsa_key.p8 -pubout -out ~/.snowflake/rsa_key.pub

# Get fingerprint for JWT claims
openssl rsa -in ~/.snowflake/rsa_key.p8 -pubout -outform DER | \
  openssl dgst -sha256 -binary | openssl enc -base64

Register Public Key with User

-- In Snowflake worksheet (requires ACCOUNTADMIN or SECURITYADMIN)
ALTER USER my_user SET RSA_PUBLIC_KEY='MIIBIjANBgkq...';

JWT Claim Format

iss: ACCOUNT_LOCATOR.USERNAME.SHA256:fingerprint
sub: ACCOUNT_LOCATOR.USERNAME

Example:

iss: NZ90655.JEZWEB.SHA256:jpZO6LvU2SpKd8tE61OGfas5ZXpfHloiJd7XHLPDEEA=
sub: NZ90655.JEZWEB

SPCS Container Authentication (v4.2.0+)

New in January 2026: Connector automatically detects and uses SPCS service identifier tokens when running inside Snowpark Container Services.

# No special configuration needed inside SPCS containers
import snowflake.connector

# Auto-detects SPCS_TOKEN environment variable
conn = snowflake.connector.connect()

This enables seamless authentication from containerized Snowpark services without explicit credentials.

Source: Release v4.2.0

Snow CLI Commands

Project Management

# Initialize project
snow init

# Execute SQL
snow sql -q "SELECT 1"
snow sql -f query.sql

# View logs
snow logs

Native App Commands