En

agent-bom ingest

Local validation and security audit of cloud resource inventory JSON

数据分析 分析云资源清单生成审计报告 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 ingest.
It is used for: Local validation and security audit of cloud resource inventory JSON
Full Skill content: https://321skill.com/skills/agent-bom-ingest-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 problem of validating cloud resource inventories and performing local security audits. In day-to-day operations, teams often need to export asset inventories from multiple cloud platforms like AWS, Azure, and GCP. However, directly handing these inventories over to third-party tools or uploading them to the cloud poses security risks. This tool allows users to perform format validation, vulnerability scanning, and policy assessment locally on the inventory file without exposing cloud credentials. It can also generate export results suitable for auditing, compliance checks, or graphical presentation.

Usage is straightforward. You simply prepare a JSON-formatted inventory file that conforms to the built-in schema, which can originate from a CMDB, an endpoint collector, or an AI Agent workflow. Then, run agent-bom mcp validate inventory.json for format validation, followed by agent-bom scan --inventory inventory.json --format json --output findings.json for local scanning. Finally, you can export the results to formats like SARIF, HTML, Markdown, or CycloneDX as needed. Using natural language instructions, an AI assistant can automatically chain these steps, for example: "Validate my AWS inventory.json and generate an HTML audit report."

It is ideally suited for operations teams, security auditors, and compliance departments that require strict control over cloud resource security. This is particularly valuable for organizations that already have a mature CMDB or endpoint collector but wish to analyze inventories without trusting third-party tools. For enterprises needing to submit audit materials to regulators, this tool can generate trusted local evidence without exposing cloud credentials.

It is recommended to use this tool for local validation each time a new inventory is exported from a cloud platform to ensure data integrity and security. Note that the tool requires the inventory to strictly comply with the packaged inventory.schema.json contract, otherwise it will error. Also, it does not automatically connect to cloud platforms; all data must be manually provided via JSON files. The optional cloud push feature requires the user to explicitly provide a target URL and authentication method. The tool includes built-in sensitive information masking to ensure credentials are not printed or leaked.

Key Features

Unlike cloud audit tools such as CloudHealth or AWS Config, agent-bom-ingest runs entirely locally, requires no cloud credentials, and supports direct input of JSON inventories from various sources like CMDBs, endpoint collectors, and AI Agents, eliminating the risk of sensitive permissions being leaked to third-party tools.

Limitations

Requires Python 3.11+ and agent-bom 0.84.4+. The inventory file must strictly comply with the packaged `inventory.schema.json` contract, otherwise validation will fail.

FAQ

Do I need to provide cloud platform access keys?

No. This tool only accepts pre-prepared JSON inventory files from the user and requires no cloud credentials. The optional push-to-control-plane operation requires an API token, but the tool strictly protects it from being leaked.

Which cloud platforms are supported?

It supports various sources including AWS, Azure, GCP, Snowflake, CMDBs, and endpoint collectors, as long as the inventory conforms to the schema.

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-ingest-x-8/raw/index.md to read the original Skill definition (Markdown format) for agent-bom ingest, and install it according to the instructions.