agent-bom discover azure
Discover Azure AI assets and generate a 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 azure. It is used for: Discover Azure AI assets and generate a standardized inventory Full Skill content: https://321skill.com/skills/agent-bom-discover-azure-x-8/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 Agent and MCP-related assets on Azure Cloud. In practical operations, teams often need to manually log into the Azure portal and inspect services one by one, which is inefficient and prone to oversight—especially when a subscription contains a mix of resources like Azure OpenAI, Container Apps, AKS, Functions, and ML Workspaces. There is a lack of a standardized inventory to support subsequent compliance audits and cost analysis.
Usage is straightforward: simply run a Python command in an environment already configured with Azure CLI or a managed identity, specifying the subscription ID and output path. It automatically calls the Azure SDK for read-only discovery, organizing AI-related assets (such as Cognitive Services, Container Apps, Function Apps, etc.) into a JSON inventory compliant with the agent-bom specification. You can then use the agent-bom scan command to perform security scans on this inventory. The entire process does not require exposing long-term Azure credentials to agent-bom.
It is well-suited for operations engineers and agent development teams that need to regularly inventory cloud-based AI infrastructure. Organizations already using agent-bom for asset management can seamlessly integrate Azure assets into a unified inventory, enabling cross-cloud (e.g., GCP, Snowflake) asset comparison and dependency analysis.
We recommend integrating this tool into CI/CD pipelines or regular inspection tasks to automatically generate daily/weekly asset snapshots. Note that it performs read-only discovery and will not modify any Azure resources. The runtime environment requires Python 3.11+ and Azure read-only credentials (Azure CLI or managed identity is recommended); avoid using long-term client secrets.
Key Features
Unlike Azure's native resource explorer or PowerShell scripts, it specifically focuses on AI Agent and MCP-related assets (such as Azure OpenAI, Container Apps, AKS, etc.) and outputs a standardized agent-bom JSON format. This output can be directly used for subsequent vulnerability scanning and dependency auditing, whereas native tools only provide unstructured resource lists.
Limitations
Requires a Python 3.11+ environment and read-only credentials such as Azure CLI, managed identity, or a service principal; direct use of long-term client secrets is not supported. Currently, it only supports Azure public cloud and does not cover Azure China or government clouds.
FAQ
How do I install this Skill?
Install the agent-bom package via pip: `pip install agent-bom`, then use the `examples/operator_pull/azure_inventory_adapter.py` script from the repository. You can also clone and run it directly from GitHub.
What Azure permissions are required?
Read-only permissions (Reader role) for the Azure subscription are required. It is recommended to use Azure CLI login, a managed identity, or short-term service principal credentials; do not use client secrets.
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-azure-x-8/raw/index.md to read the original Skill definition (Markdown format) for agent-bom discover azure, and install it according to the instructions.
Raw Markdown URL for AI: /skills/agent-bom-discover-azure-x-8/raw/index.md