Skip to content
Get packs

Gemini API on Agent Platform

Guides agents to build with the Gemini API and Google Gen AI SDKs on Google Cloud's Agent Platform (formerly Vertex AI).

Source Download skill
Repository
google/skills
Stars
21,782
Licence
Apache-2.0
Files
11 files

What it does

Once installed, the agent knows how to call Google's Gemini models through Google Cloud's enterprise Agent Platform (previously called Vertex AI). It steers the agent toward the current Gen AI SDKs for Python, JavaScript/TypeScript, Go, Java and C#, explains how to set up Google Cloud credentials or an API key, picks recent model names, and provides reference notes on features such as multimodal input, function calling, structured output, caching, embeddings, media generation, safety settings and the Live realtime API. Many detailed guides are in separate reference files the agent reads when needed.

Great for

  • Developers building apps on Google Cloud who want Gemini code written against the current SDKs
  • Enterprise teams already using Google Cloud projects, credentials or Agent Platform
  • Engineers migrating away from the deprecated google-cloud-aiplatform, @google-cloud/vertexai or google-generativeai libraries

Skip it if

  • Personal users with only a consumer Gemini account and no Google Cloud project, since the skill requires Cloud credentials or an Express Mode API key
  • Projects that must pin specific model versions without review, because the skill's model list may change and includes preview models that could be unstable or costly
  • Anyone expecting the skill to run code or verify results itself, since it only gives the agent instructions and reference material

Try asking

  • “Write a Python script that summarises a PDF using Gemini on Agent Platform”
  • “Add Gemini function calling to my Node.js app using the @google/genai SDK”
  • “Set up Google Cloud authentication so my Go service can call the Gemini API”
  • “Migrate my code from google-generativeai to the new Google Gen AI SDK”
Step-by-step guide: how to use Gemini API on Agent PlatformInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 11 original files, the Apache-2.0 licence and an attribution note.

    gemini-api.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip gemini-api.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip gemini-api.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Set GOOGLE_CLOUD_PROJECT and related environment variables rather than pasting API keys into code, and never commit keys to a repository
  • Ask for a specific region only when you need one; the default global endpoint routes automatically
  • Check the model names against current Google documentation before production use, since the skill itself advises consulting stable versions

What's inside

  • references/advanced_features.md
  • references/attribution.md
  • references/bounding_box.md
  • references/embeddings.md
  • references/live_api.md
  • references/media_generation.md
  • references/model_tuning.md
  • references/safety.md
  • references/structured_and_tools.md
  • references/text_and_multimodal.md
  • SKILL.md

Redistributed unmodified from google/skills under the Apache-2.0 licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.