agent-bom discover snowflake
Scan Snowflake AI Assets to Generate a Standard 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: Scan Snowflake AI Assets to Generate a Standard Inventory Full Skill content: https://321skill.com/skills/agent-bom-discover-snowflake-x-10/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 unifying the inventory of AI assets within Snowflake environments. In practical operations, teams often need to manually query the distribution of components like Cortex, Snowpark, Streamlit, and MCP, lacking a standardized inventory format, which complicates subsequent audits and migrations. The tool automatically discovers assets using read-only credentials and outputs a standardized agent-bom JSON, providing a reliable foundation for infrastructure management.
Usage is straightforward: ensure Python 3.11+ and agent-bom (with the snowflake extra) are installed locally, and have Snowflake read-only credentials ready (SSO, OAuth, or key pair authentication is recommended). You can then trigger the scan using natural language instructions, such as "Scan my Snowflake environment and generate an AI asset inventory." The tool connects to the specified account, traverses resources like Cortex, Snowpark, Notebook, and Streamlit, and writes the results to a JSON file you specify. Throughout the process, credentials remain only in the operational environment and are not leaked to third parties.
It is well-suited for DevOps engineers and agent development teams who need unified management of Snowflake AI infrastructure. Organizations already using the agent-bom ecosystem for inventory management can seamlessly integrate this Skill to achieve a complete workflow from discovery to scanning. For scenarios requiring regular audits of Snowflake resources to ensure compliance, this tool can significantly reduce the manual effort of inventorying.
It is recommended to run in discover-only mode during the first use to verify the completeness and accuracy of the output JSON. Note that the tool primarily relies on read-only permissions and will not modify any Snowflake resources. If MFA is enabled on the account, using key pair authentication is advised to avoid interactive input. Additionally, the output file path must be specified by the operator; the tool does not automatically upload or persist data.
Key Features
Unlike directly using Snowflake SQL queries, it outputs a standardized agent-bom inventory JSON that can be directly consumed by other tools in the agent-bom ecosystem. It also enforces the use of read-only credentials and does not store long-term credentials, whereas native queries often require manual handling of credentials and format conversion.
Limitations
Requires Python 3.11+, agent-bom installed with the snowflake extra, and only supports Snowflake read-only credentials (SSO/OAuth/key pair recommended; password authentication is not supported). The scan scope is limited by account permissions and network reachability.
FAQ
How do I configure Snowflake credentials?
It is recommended to set environment variables SNOWFLAKE_ACCOUNT, SNOWFLAKE_USER, SNOWFLAKE_AUTHENTICATOR (e.g., snowflake_jwt), and SNOWFLAKE_PRIVATE_KEY_PATH, or use the configuration file ~/.snowflake/connections.toml. Avoid using passwords directly.
Which Snowflake objects are supported for discovery?
Supports Cortex AI, Snowpark, Notebook, Streamlit, MCP, and AI observability-related assets, including warehouses, functions, stored procedures, streams, tasks, and more.
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-10/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-10/raw/index.md