agent-bom discover aws
Inventory AWS AI Assets and Generate Standardized Bill of Materials
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 AWS AI Assets and Generate Standardized Bill of Materials Full Skill content: https://321skill.com/skills/agent-bom-discover-aws-x-3/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 inventorying AI Agent and MCP-related assets within an AWS cloud environment. Operations and development personnel often need to understand what resources exist in their account, such as Bedrock Agents, ECS services, SageMaker endpoints, Lambda functions, EKS clusters, Step Functions state machines, EC2 instances, and more. However, manually traversing the console or writing scripts is time-consuming and results in inconsistent output formats. This Skill automatically discovers these resources via read-only AWS SDK calls and generates a JSON inventory compliant with the agent-bom specification, providing a standardized data foundation for subsequent security scanning, cost analysis, and compliance audits.
Usage is straightforward. You simply need to ensure Python 3.11+ and the agent-bom library are installed locally, and AWS credentials are configured (it's recommended to use short-term credentials obtained via AWS SSO, WebIdentity, or STS). You can then trigger a scan using natural language instructions (e.g., "Inventory all AI-related assets in the current AWS account"). The Skill will call AWS STS to verify identity, then sequentially query resources from services like Bedrock, ECS, SageMaker, Lambda, EKS, Step Functions, and EC2, ultimately writing the inventory to a JSON file you specify. You have full control over whether to pass this inventory to agent-bom for deep scanning.
It is well-suited for operations engineers and agent developers managing multi-cloud AI infrastructure. Particularly for teams already using agent-bom for full lifecycle management of AI assets, this Skill enables rapid acquisition of standardized AWS-side inventories, unifying the format with inventories from platforms like GCP and Azure to achieve cross-cloud asset visibility and compliance auditing. For security auditors, it can also help identify unauthorized AI service deployments, reducing security risk.
It is recommended to first use the agent-bom trust command to confirm the required scope of read-only IAM permissions and ensure credentials are short-lived. Note that this Skill performs discovery operations only; it will not modify any AWS resources or send credentials or data to third-party services. The generated JSON inventory automatically redacts fields that may contain credential patterns. For regular inventory needs, you can integrate it into CI/CD pipelines, using scheduled tasks to automatically generate reports.
Key Features
Unlike native AWS resource inventory tools (such as AWS Config or Resource Explorer), it specifically focuses on AI Agent and MCP-related assets (Bedrock, ECS, SageMaker, etc.) and outputs standardized JSON compliant with the agent-bom specification, facilitating unified cross-cloud management and subsequent scanning and analysis.
Limitations
Requires a Python 3.11+ environment, depends on the agent-bom library (installable from PyPI), and the operator must possess AWS read-only permissions (recommended to use SSO or STS short-term credentials; long-term access keys are not supported).
FAQ
What AWS permissions are required?
Read-only permissions are required. You can view the specific list of required IAM actions via the `agent-bom trust` command. It is recommended to use AWS SSO or STS to obtain short-term credentials.
Which AWS services are supported?
Supports AI and MCP-related services such as Bedrock Agent, ECS, SageMaker, Lambda, EKS, Step Functions, and EC2. Services can be enabled or disabled via configuration.
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-3/raw/index.md to read the original Skill definition (Markdown format) for agent-bom discover aws, and install it according to the instructions.
Raw Markdown URL for AI: /skills/agent-bom-discover-aws-x-3/raw/index.md