snowflake-create-iceberg-table

SkillDatabases & data

Consult Snowflake CREATE ICEBERG TABLE parameter reference before generating any CREATE ICEBERG TABLE DDL.

Available today. Use it from your connected AI after setup.

No other account needed.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the snowflake-create-iceberg-table skill

What this skill tells your AI

The instructions your AI receives, as published by gyrus-dev/frosty in skills/snowflake-create-iceberg-table/SKILL.md and read by ahel’s review.

Before writing a CREATE ICEBERG TABLE statement:

Read references/parameters.md to review all available parameters and their defaults.
For each parameter, decide whether the user's request implies a non-default value.
Include only the parameters that differ from the default or that the user explicitly requested.
Never use CREATE OR REPLACE — always use CREATE ICEBERG TABLE IF NOT EXISTS where supported.
Always specify EXTERNAL_VOLUME and BASE_LOCATION when Snowflake is the catalog (the most common case); omit them only when integrating an external catalog.
CATALOG = 'SNOWFLAKE' is the default; omit it unless you are referencing an external catalog integration.
STORAGE_SERIALIZATION_POLICY = COMPATIBLE is the default; use OPTIMIZED only when the user explicitly wants Snowflake-optimised encoding and compression and does not need to read the files with other engines.
Note that NOT NULL and UNIQUE constraints on PRIMARY KEY columns are represented in Iceberg metadata as identifier fields but are not enforced by Snowflake at DML time.
Do not add CLUSTER BY unless the table is expected to be very large or the user requests it.
Use PARTITION BY only when the user specifies a partitioning strategy; choose identity, bucket, truncate, or temporal transforms as appropriate.

Advanced