agent-bom discover snowflake
Automatically inventory Snowflake AI assets and generate standardized BOMs.
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: Automatically inventory Snowflake AI assets and generate standardized BOMs. Full Skill content: https://321skill.com/skills/agent-bom-discover-snowflake-x-2/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 comprehensively inventorying AI assets in enterprise Snowflake environments. In practice, operations and agent teams often need to understand what Cortex services, Snowpark jobs, Notebooks, Streamlit applications, and MCP tools are deployed within Snowflake, but manual discovery is inefficient and prone to omissions. This tool automatically discovers these assets using read-only credentials and outputs a standardized agent-bom JSON, providing a reliable foundation for subsequent asset management and auditing.
Usage is straightforward: you only need to install Python 3.11+ and the agent-bom library (with the snowflake extension) locally, then configure read-only Snowflake credentials (SSO, OAuth, or key-pair authentication is recommended). You can then trigger the discovery process via the command line or natural language instructions. The script connects to your Snowflake account, scans all accessible resources like Cortex and AI Observability, and writes the results to a JSON file you specify. The entire process does not require handing credentials to any third-party service, keeping data security under your control.
It is well-suited for operations engineers, data architects, and agent developers who need to maintain Snowflake asset inventories. This is especially true for teams that have already deployed Snowflake Cortex AI services and wish to incorporate them into standardized bill-of-materials management. Whether for daily inspections, compliance audits, or providing infrastructure context to AI agents, this tool can significantly improve efficiency.
Before first use, ensure your Snowflake account has sufficient read-only permissions (e.g., the MONITOR role) and prioritize using short-lived OAuth tokens or key-pair authentication to avoid long-term password exposure. Note that it primarily focuses on Snowflake AI and Cortex-related assets; it does not inventory traditional database tables, views, or other non-AI resources. For full coverage, please combine it with other tools.
Key Features
Unlike Snowflake's native INFORMATION_SCHEMA queries, this tool not only discovers basic objects like tables and views but also identifies AI-level assets such as Cortex AI services, Snowpark jobs, Notebooks, Streamlit applications, and MCP connections. It outputs in the standardized agent-bom format, facilitating integration with multi-tool chains.
Limitations
Requires Python 3.11+ and the snowflake extension for agent-bom. It depends on read-only Snowflake credentials provided by the operator and can only discover assets accessible to the current account. Inventorying assets across accounts or organizations requires additional configuration.
FAQ
Do I need to provide my Snowflake password to this tool?
No. The tool prioritizes using SSO, OAuth, or key-pair authentication to avoid cleartext password transmission. Credentials are only stored in the local execution environment.
Can the generated JSON file be used directly by other tools?
Yes. The output adheres to the agent-bom specification and can be consumed by any scanner or asset management tool that supports the agent-bom format.
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-2/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-2/raw/index.md