agent-bom discover snowflake
Auditing Snowflake AI Infrastructure Assets
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: Auditing Snowflake AI Infrastructure Assets Full Skill content: https://321skill.com/skills/agent-bom-discover-snowflake-x/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 unified discovery and auditing of AI assets within a Snowflake environment. In practice, data teams and operations personnel often need to inventory AI-related resources such as Cortex, Snowpark, Streamlit, and MCP, but lack standardized aggregation tools. This leads to incomplete asset lists, inconsistent formats, and difficulties in integrating with other systems.
Usage is straightforward. First, install agent-bom and enable the snowflake extension, then configure read-only Snowflake credentials (SSO, OAuth, or key pair authentication is recommended). Next, run the provided Python script. It will automatically connect to Snowflake, scan assets related to Cortex, Snowpark, Notebooks, Streamlit, MCP, and AI observability, and finally generate a JSON inventory file compliant with the agent-bom specification, saving it to your specified path.
It is well-suited for data analysts, operations engineers, and agent developers who need to manage Snowflake AI infrastructure. This is particularly valuable for teams or individuals already using Snowflake as a data platform and wishing to establish a unified AI asset inventory, conduct compliance audits, or enable cross-tool integration.
It is recommended to use a read-only role during initial deployment to ensure no Snowflake resources are modified. Note that it primarily relies on a Python 3.11+ environment, and credential information does not leave the operator's environment, ensuring high security. The generated JSON inventory can subsequently be scanned or pushed via agent-bom.
Key Features
Unlike Snowflake's native INFORMATION_SCHEMA or ACCOUNT_USAGE views, it outputs a standardized agent-bom inventory, supports unified asset discovery across tools (e.g., MCP, Cortex), and does not require providing long-term credentials to agent-bom—credentials are used only locally by the operator.
Limitations
Requires Python 3.11+, installation of the agent-bom snowflake extension, and only supports read-only Snowflake credentials controlled by the operator; it does not support write or modify operations.
FAQ
What Snowflake permissions are required?
The operator must provide read-only Snowflake credentials. Using SSO, OAuth, or key pair authentication is recommended to avoid direct password usage.
Will it modify data in Snowflake?
No, this Skill is discovery-only. It performs only queries and API calls and does not create, modify, or delete any Snowflake resources.
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/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/raw/index.md