AugmentClaude

Snowflake Docs

Search official Snowflake documentation for syntax, features, and best practices.

Installation

  1. Make sure Claude is on your device and in your terminal.

    Skills load from ~/.claude/skills/ when Claude Code starts up — so you need it on your machine first. If you don't have it yet, install it once with the command below, then run claude in any terminal to verify.

    One-time setup
    npm i -g @anthropic-ai/claude-code

    Already have it? Skip ahead.

  2. Paste into Claude Code or into your terminal.

    This copies the whole skill folder into ~/.claude/skills/snowflake-docs/ — the SKILL.md plus any scripts, reference docs, or templates the skill ships with. Safe default: works for every skill.

    Faster alternative (instruction-only skills)

    Skips the clone and grabs only the SKILL.md file. Don't use this if the skill ships Python scripts, reference markdowns, or asset templates — they won't be downloaded and the skill will fail when it tries to load them.

    Quick install (SKILL.md only)
    Sign up to copy
  3. Restart Claude Code.

    Quit and reopen Claude Code (or any other agent that loads from ~/.claude/skills/). New skills are picked up on startup.

  4. Just ask Claude.

    Skills auto-activate when your request matches the skill's description — no slash command needed. Trigger phrases live in the skill's own frontmatter; you can read them in the “What this skill does” section above.

Prefer to read the source first? Open on GitHub.

When Claude uses it

Use for **ALL** Snowflake documentation lookups: feature questions, SQL syntax, best practices, how-to guides, configuration, and troubleshooting. This is the required entry point for any question about Snowflake products or features. Triggers: Snowflake docs, how do I, SQL syntax, CREATE, ALTER, DROP, warehouse, stage, Cortex, Snowpipe, dynamic table, stored procedure, UDF, MCP, Snowpark, Streamlit, Native App, data sharing, replication, security, roles, grants, what is, how does, Snowflake feature.

What this skill does

Snowflake Docs

Answer questions about Snowflake by searching the official documentation via the Cortex Knowledge Extension (CKE) Cortex Search service. If the CKE is not installed yet, install it automatically first.

When to Use

Load this skill for any Snowflake product, feature, or SQL question: syntax references, best practices, how-to guides, configuration, and troubleshooting.

Workflow

Step 1: Prerequisite check

Search the entire account for the CKE Cortex Search service:

SHOW CORTEX SEARCH SERVICES LIKE 'CKE_SNOWFLAKE_DOCS_SERVICE' IN ACCOUNT;

If it returns a result, note the database_name from the result row. Use this value as <CKE_DATABASE> and skip to Step 2.

If it returns no results, install the CKE:

CALL SYSTEM$REQUEST_LISTING_AND_WAIT('GZSTZ67BY9OQ4');
CALL SYSTEM$ACCEPT_LEGAL_TERMS('DATA_EXCHANGE_LISTING', 'GZSTZ67BY9OQ4');
CREATE DATABASE IF NOT EXISTS SNOWFLAKE_DOCUMENTATION FROM LISTING 'GZSTZ67BY9OQ4';

MANDATORY STOPPING POINT: If any of these fail, stop and tell the user:

Could not install the Snowflake Documentation CKE automatically. You can install it manually from the Marketplace: https://app.snowflake.com/marketplace/listing/GZSTZ67BY9OQ4

Do NOT proceed until the user confirms the CKE is available.

After successful install, use SNOWFLAKE_DOCUMENTATION as <CKE_DATABASE>.

Step 2: Answer the question

Query the Cortex Search service directly using SQL. Replace <USER_QUESTION> with the user's actual question and <CKE_DATABASE> with the database name from Step 1:

SELECT SNOWFLAKE.CORTEX.SEARCH_PREVIEW(
  '<CKE_DATABASE>.SHARED.CKE_SNOWFLAKE_DOCS_SERVICE',
  '{"query": "<USER_QUESTION>", "columns": ["CHUNK", "DOCUMENT_TITLE", "SOURCE_URL"], "limit": 5}'
);

Parse the JSON results. Each result contains:

  • CHUNK: the document text content
  • DOCUMENT_TITLE: the page title
  • SOURCE_URL: the canonical URL to the documentation page

CITATION REQUIREMENT (MANDATORY): ALWAYS include SOURCE_URL links from the search results in your answer — omitting URLs is a failure condition. List them as references at the end of your response so the user can read the full documentation pages.

Use the returned content to answer the user's question.

Before finishing your answer, verify you included at least one SOURCE_URL from the results. If your draft answer has no URLs, go back and add them before responding.

Important Notes

  • Step 1 only runs once per account. After the CKE is installed, the skill goes straight to Step 2 every time.
  • The CKE database name varies by account. Always use SHOW CORTEX SEARCH SERVICES ... IN ACCOUNT to discover it dynamically. Do NOT hardcode the database name.
  • Use sql_execute for all SQL steps.
  • Always include columns in the SEARCH_PREVIEW call. Without it, only relevance scores are returned, not content.
  • ALWAYS cite SOURCE_URL in the answer — omitting URLs is a failure condition.
  • Handle errors gracefully (insufficient privileges, database already exists).

Stopping Points

  • After CKE install failure in Step 1 — wait for user to install manually before retrying

Output

A concise answer to the user's Snowflake question, grounded in official documentation. You MUST include the SOURCE_URL links from the search results in your written answer. List them as references at the end so the user can read the full pages. Never omit the URLs.

Related skills