En

agent-bom discover aws

Inventory AI Agents and MCP Assets on AWS

开发流程 盘点AWS云上AI资产生成清单 通用 ★ 2.7k 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 aws.
It is used for: Inventory AI Agents and MCP Assets on AWS
Full Skill content: https://321skill.com/skills/agent-bom-discover-aws-x-6/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 quickly discovering and inventorying AI agent and MCP-related assets within an AWS cloud environment. In practical operations or security audits, teams often need to understand what Bedrock agents, ECS services, SageMaker endpoints, Lambda functions, EKS clusters, Step Functions state machines, EC2 instances, and other resources are deployed in their AWS accounts. Manually querying each resource is time-consuming and prone to omissions. This tool automatically scans these resources and outputs a standardized agent-bom inventory JSON, providing a unified data foundation for subsequent dependency analysis, security scanning, and compliance audits.

Usage is straightforward. You only need to ensure Python 3.11+ and the agent-bom package are installed locally, and that AWS credentials are configured (using AWS SSO, WebIdentity, or short-term credentials generated via STS is recommended). You can then trigger the tool with a natural language instruction (e.g., "Inventory all AI-related resources in my AWS account"). The tool will call the AWS SDK for read-only queries and write the results to a JSON file you specify. The entire process does not require handing long-term credentials to agent-bom; the data flow is entirely controlled by the operator, who can choose whether to pass the inventory to agent-bom for further scanning.

It is well-suited for operations engineers, agent developers, and security auditors managing multi-cloud AI infrastructure. This is especially true for teams already using agent-bom for dependency management and security scanning, as they can use this tool to quickly generate standardized asset inventories, bridging the complete workflow from resource discovery to risk analysis. For organizations needing to regularly audit AWS AI resource usage, this Skill can significantly improve inventory efficiency and accuracy.

It is recommended to first use the agent-bom trust command before initial use to confirm the required scope of read-only IAM permissions and ensure the AWS credential chain is correctly configured. Note that it primarily relies on the AWS SDK's read-only APIs and will not modify any existing resources; credential pattern values are automatically redacted in the generated inventory JSON. If an account contains a large number of resources, scanning region by region is advised to avoid API throttling.

Key Features

Unlike AWS Config or AWS Resource Explorer, this tool specifically focuses on AI agent and MCP-related assets (such as Bedrock, SageMaker, ECS, etc.) and outputs an inventory in the standard agent-bom format. This allows for seamless integration with agent-bom's scanning, policy, and export functionalities, whereas general-purpose resource discovery tools cannot directly generate this type of structured inventory.

Limitations

Requires a Python 3.11+ environment, depends on the agent-bom package (to be installed from PyPI or GitHub), and only supports access via the AWS SDK credential chain; it does not support directly passing long-term access keys.

FAQ

What permissions does this Skill require?

It requires AWS read-only permissions. It is recommended to use the `agent-bom trust` command to view the minimum required IAM policy and to prioritize using short-term credentials (SSO/STS).

How do I use the generated JSON file?

It can be directly passed to agent-bom for scanning, or archived as an asset inventory, or imported into other tools for analysis.

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-aws-x-6/raw/index.md to read the original Skill definition (Markdown format) for agent-bom discover aws, and install it according to the instructions.