agent-bom discover gcp
One-click inventory of GCP AI assets, generating a standardized manifest.
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 gcp. It is used for: One-click inventory of GCP AI assets, generating a standardized manifest. Full Skill content: https://321skill.com/skills/agent-bom-discover-gcp-x-2/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 inventorying AI assets on Google Cloud Platform (GCP). In practice, operations and AI agent development teams often need to quickly understand which Vertex AI models, Cloud Run services, Cloud Functions, GKE clusters, and other resources related to AI Agents and MCP are deployed in the current GCP project. However, manually querying each resource via gcloud commands is inefficient and prone to oversight.
Usage is straightforward. In an environment with Python 3.11+ and the agent-bom library installed, simply instruct the AI using natural language to "Inventory the AI assets in the current GCP project." It will automatically invoke the GCP SDK (using your Application Default Credentials or Workload Identity) for a read-only scan and output the results as a standardized agent-bom inventory JSON file. You can then perform local analysis on this JSON or push it to other systems.
It is ideal for operations engineers and AI agent development teams that require regular audits of AI infrastructure on GCP. Organizations already using agent-bom as a unified asset inventory standard can integrate it seamlessly into existing workflows without additional configuration for long-term credentials.
It is recommended to ensure GCP ADC (Application Default Credentials) or Workload Identity is configured before first use. Note that this Skill performs read-only operations only, does not modify any GCP resources, and does not leak credential information to external services. The output path is user-specified, and centralized management is advised.
Key Features
Unlike manually querying with gcloud commands one by one, it automatically discovers all GCP resources related to AI Agents and MCP (Vertex AI, Cloud Run, Cloud Functions, GKE) and outputs a standardized agent-bom JSON format. It does not require providing long-term GCP credentials to agent-bom, using only the operator's existing short-term credentials.
Limitations
Requires a Python 3.11+ environment, depends on the agent-bom library (installable from PyPI), supports only the GCP platform, and requires the operator to have GCP read-only permissions (ADC/Workload Identity/short-term service account).
FAQ
What GCP permissions are required?
Only read-only permissions are required. It is recommended to use Application Default Credentials or Workload Identity; long-term service account keys are not needed.
What format is the output file?
The output is a standardized agent-bom inventory JSON file containing information for all discovered AI Agent and MCP-related assets.
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-gcp-x-2/raw/index.md to read the original Skill definition (Markdown format) for agent-bom discover gcp, and install it according to the instructions.
Raw Markdown URL for AI: /skills/agent-bom-discover-gcp-x-2/raw/index.md