En

agent-bom discover azure

Scan Azure environments to generate an AI asset inventory.

开发流程 扫描Azure环境生成AI资产清单 通用 ★ 2.6k Updated 2026-08-02

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: Scan Azure environments to generate an AI asset inventory.
Details: https://321skill.com/en/skills/agent-bom-discover-azure/

This skill addresses the challenge of quickly discovering and inventorying AI Agent and MCP-related assets within Azure cloud environments. Operations personnel and agent developers often need to manually log into the Azure portal or use the CLI to query services one by one, which is inefficient and prone to oversight. This is especially true when dealing with numerous subscriptions and resource groups, making it difficult to form a unified asset view.

Usage is straightforward: you simply run a single command in an authenticated Azure CLI environment or use a natural language instruction for an AI agent to execute the scan. It automatically calls the Azure Resource Graph API, traverses all subscriptions, identifies assets such as Azure AI Studio, Cognitive Services, Bot Service, and MCP servers, and outputs a JSON inventory compliant with the agent-bom specification. This inventory can subsequently be used for security scanning without granting agent-bom long-term Azure credentials, utilizing only temporary tokens.

It is ideal for operations engineers managing multi-cloud AI infrastructure, agent development teams, and security auditors. Teams that have deployed multiple Azure AI services but lack a unified asset inventory and compliance checks can especially benefit by quickly gaining a comprehensive overview of their assets.

We recommend using this tool uniformly during periodic asset inventory checks or security audits. Note that it primarily relies on the Azure Resource Graph, so ensure the execution environment has sufficient read permissions (e.g., the Reader role) and that the scan scope is limited to the currently logged-in Azure tenant and subscriptions.

Key Features

Unlike Azure Resource Graph Explorer or manually writing scripts, it is specifically designed for the structured discovery of AI Agent and MCP assets and directly outputs a standardized agent-bom JSON without requiring additional conversion. It also mitigates the risk of long-term credential exposure through a temporary token mechanism.

Limitations

Requires the Azure CLI to be logged in with at least Reader permissions, depends on the availability of the Azure Resource Graph API, and does not support preview AI services in all Azure regions.

FAQ

Does agent-bom require long-term Azure credentials?

No, it uses temporary tokens from the current Azure CLI session, and the credentials expire after the scan completes.

How can the generated JSON inventory be used?

It can be directly imported into other tools within the agent-bom ecosystem for security scanning, compliance auditing, or asset visualization.