<< All versions

Skill v1.0.0

currentAutomated scan100/100
gyrus-dev/frosty/snowflake-create-compute-pool
──Details
PublishedNovember 3, 2026 at 05:43 AM
Content Hashsha256:49fd9a34f9ec8094...
Git SHA56db724d9dcc
──Files
Files (1 file, 1.1 KB)
SKILL.md1.1 KBactive
SKILL.md · 14 lines · 1.1 KB

name: snowflake-create-compute-pool description: Consult Snowflake CREATE COMPUTE POOL parameter reference before generating any CREATE COMPUTE POOL DDL. metadata: version: "1.0.0"


Before writing a CREATE COMPUTE POOL statement:

  1. Read references/parameters.md to review all available parameters and their defaults.
  2. For each parameter, decide whether the user's request implies a non-default value.
  3. Include only the parameters that differ from the default or that the user explicitly requested.
  4. Never use CREATE OR REPLACE — always use CREATE COMPUTE POOL IF NOT EXISTS.
  5. Always set MIN_NODES, MAX_NODES, and INSTANCE_FAMILY — these are required. Choose INSTANCE_FAMILY based on workload: CPU_X64_XS/S/M/L for general workloads, HIGHMEM_X64_S/M/L for memory-intensive workloads, GPU_NV_S/M/L for GPU workloads. Set INITIALLY_SUSPENDED = TRUE for pools that should not incur cost immediately. Set AUTO_SUSPEND_SECS conservatively (default 3600) to avoid runaway costs. Use FOR APPLICATION only when the pool is dedicated to a Snowflake Native App.
All versions