Apr 15, 2026: Snowflake documentation for AI agents and LLMs¶
Snowflake documentation is now easier for AI coding assistants, agents, and large language models (LLMs) to discover and consume.
Hierarchical llms.txt
docs.snowflake.com/llms.txt now follows a hierarchical structure. Instead of one large file containing every page, the root file links to section-level indexes (for example, SQL Commands). Agents and tools can fetch only the sections they need, reducing token usage and improving relevance.
Markdown versions of every page
Every documentation page is also available in Markdown by appending .md to the URL. For
example:
CREATE TABLE has a corresponding Markdown page create-table.md.
Virtual warehouses has a corresponding Markdown page warehouses-overview.md.
Markdown is smaller than the equivalent HTML and strips away navigation, scripts, and other elements that aren’t useful for LLMs. Tools such as Cortex Code, Cursor, and Claude Code can use these URLs directly as context.