agent-bom discover snowflake
Discover Snowflake AI assets and generate standardized inventory
Install & Use
Copy this prompt and send it to your AI assistant (Claude / Cursor / TRAE / Codex / WorkBuddy etc.) to auto-install:
Help me install this AI Skill: agent-bom discover snowflake. It is used for: Discover Snowflake AI assets and generate standardized inventory Full Skill content: https://321skill.com/skills/agent-bom-discover-snowflake-x-11/raw/index.md Read that page and install it.
The prompt includes a link to the full Skill content. You can also view the full content.
This Skill addresses the challenge of uniformly inventorying AI assets within Snowflake environments. In practice, enterprises often use Snowflake Cortex, Snowpark, Streamlit, MCP, and various AI observability components simultaneously, but lack a standardized method to discover and document these assets. Manual tracking is not only time-consuming and prone to omissions, but also fails to produce a standardized inventory consumable by downstream tools.
Usage is straightforward. You only need read-only access to Snowflake (SSO, OAuth, or key-pair authentication is recommended) and ensure Python 3.11+ and agent-bom (with the snowflake extension) are installed locally. Then, a single command triggers the discovery process: it automatically connects to your Snowflake account, queries AI-related resources like Cortex, Snowpark, Notebooks, Streamlit, and MCP, and outputs a JSON inventory file compliant with the agent-bom specification. The entire process does not require exposing long-term credentials to the agent-bom service; credentials remain solely in the operator's environment.
It is ideal for teams or individuals needing unified management of Snowflake AI infrastructure. This is especially true for enterprises that have deployed multiple AI components (e.g., Cortex inference, Snowpark pipelines, Streamlit applications) or require regular generation of asset inventories for auditing, cost analysis, or migration planning. Operations engineers, data analysts, and agent developers can all benefit.
It is recommended to configure a Snowflake read-only role before first use and ensure network access to the corresponding Snowflake account endpoint. Note that this Skill only performs discovery and will not modify any Snowflake resources. The generated JSON inventory can be further scanned by agent-bom or pushed to other systems. If using password authentication, avoid entering the password directly in chat; prioritize key-pair or OAuth.
Key Features
Unlike directly using native Snowflake queries (e.g., INFORMATION_SCHEMA), this Skill automatically generates a standardized agent-bom inventory JSON. Crucially, it does not require granting long-term credentials to the agent-bom service; all credentials are retained only in the operator's local environment, offering higher security.
Limitations
Requires Python 3.11+ and installation of the agent-bom snowflake extension (`pip install "agent-bom[snowflake]"`). Must use Snowflake read-only credentials (SSO/OAuth/key-pair recommended; direct password input in chat is not supported).
FAQ
Are any additional tools required for installation?
Yes, you need to install agent-bom with the snowflake extension: `pip install "agent-bom[snowflake]"`.
How do I configure Snowflake credentials?
Prefer using environment variables such as SNOWFLAKE_ACCOUNT, SNOWFLAKE_USER, SNOWFLAKE_AUTHENTICATOR, etc., or configure via `~/.snowflake/connections.toml`. Avoid pasting passwords or private keys directly in chat.
Installation guide for AI assistants
If your AI coding assistant (Claude Code, Cursor, TRAE etc.) can see this page, send it this message to auto-install:
Visit https://321skill.com/skills/agent-bom-discover-snowflake-x-11/raw/index.md to read the original Skill definition (Markdown format) for agent-bom discover snowflake, and install it according to the instructions.
Raw Markdown URL for AI: /skills/agent-bom-discover-snowflake-x-11/raw/index.md