Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add jpantsjoha/googlecloud-plugin --skill vertex-aigit clone --depth 1 https://github.com/jpantsjoha/googlecloud-pluginWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/jpantsjoha/googlecloud-plugin/vertex-ai)<a href="https://agentmods.dev/skills/jpantsjoha/googlecloud-plugin/vertex-ai"><img src="https://agentmods.dev/badge/skills/jpantsjoha/googlecloud-plugin/vertex-ai/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/jpantsjoha/googlecloud-plugin/vertex-ai"><img src="https://agentmods.dev/badge/skills/jpantsjoha/googlecloud-plugin/vertex-ai.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00136 | $0.00876 |
| Opus 5.5 | $0.00054 | $0.00350 |
| Sonnet 5.5 | $0.00027 | $0.00175 |
| Haiku 4.5 | $0.00014 | $0.00088 |
Grade A, and why
vertex-ai scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Vertex AI
Unified ML platform for training, deploying, and serving models — including Gemini, generative AI, and custom models.
Naming currency: the platform was renamed to the Gemini Enterprise Agent Platform (GEAP) at Next '26 (2026-04-22). The rebrand is branding-only —
vertex-aipersists in gcloud groups, SDK imports, API hostnames, Terraform resources, and doc URLs, so this skill keeps them. For agentic system design (ADK, Agent Runtime, multi-agent, MCP/A2A/AP2, agent evaluation), route toagent-architect; this skill covers platform/model mechanics.
Key Products
| Product | Use |
|---|---|
| Gemini API on Vertex | Production-grade Gemini access with data governance |
| Model Garden | Browse and deploy 150+ models (Gemini, Llama, etc.) |
| Custom Training | Train on managed infrastructure (GPUs/TPUs) |
| Vertex AI Endpoints | Serve predictions from deployed models |
| Agent Builder | Build conversational agents on Vertex |
| Vertex AI Pipelines | Orchestrate ML workflows (Kubeflow Pipelines) |
Core Patterns
Call Gemini via Vertex (Python)
import vertexai
from vertexai.generative_models import GenerativeModel
vertexai.init(project="PROJECT_ID", location="us-central1")
model = GenerativeModel("gemini-2.0-flash-001")
response = model.generate_content("Explain Cloud Run in one sentence.")
print(response.text)
Deploy an endpoint (billable — confirm first)
# Billable: endpoints charge per node-hour when deployed
gcloud ai endpoints create \
--display-name=ENDPOINT_NAME \
--region=REGION \
--project=PROJECT_ID
Grant Vertex AI access (least-privilege)
gcloud projects add-iam-policy-binding PROJECT_ID \
--member="serviceAccount:[email protected]" \
--role="roles/aiplatform.user"
Safety Rules
- Endpoints incur cost even when idle — undeploy when not needed
- Use
roles/aiplatform.userfor inference;roles/aiplatform.adminonly for platform admins - Cost warning: GPUs and TPUs are expensive — confirm instance type and count before training
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 73 lines · 136 tokens per session scan A ded0836742ce
vertex-ai is a skill published in the GitHub repository jpantsjoha/googlecloud-plugin (4 stars, last pushed 2mo ago), licensed MIT. It adds 136 tokens to every session and 876 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
prompt-cache-agent-harness
Plan and inspect prompt-cache behavior for long-running Claude agent loops. Use when a user wants to split stable tool, system, and history context into cacheable layers, compare captured cache metadata, estimate cost impact from supplied pricing inputs, or keep durable memory outside the cached prefix.
structured-output
Force orxhestra agents to return typed Pydantic objects instead of free-form text using outputschema.
prompt-architect
Convert any rough, casual, or unstructured human prompt into a clean, professional, well-architected prompt. Use this whenever a user pastes a messy instruction, a brain-dump, a bullet list, or a vague request and wants it rewritten as a structured prompt — for an AI agent, a coding assistant, a writing task, or a…
qzcli
A command-line tool for managing GPU compute jobs on the Qizhi platform. GPU compute jobs use graphics processors to run demanding workloads such as machine-learning training.
obliteratus
OBLITERATUS: abliterate LLM refusals (diff-in-means).
qdrant-vector-search
High-performance vector similarity search engine for RAG and semantic search. Use when building production RAG systems requiring fast nearest neighbor search, hybrid search with filtering, or scalable vector storage with Rust-powered performance.