google-gemini/gemini-skills

Skills for the Gemini API, SDK and model/agent interactions

What it solves

It addresses the knowledge gap in LLMs caused by their fixed training data. Because software development evolves rapidly, models often lack awareness of the latest SDK changes, new libraries, and current best practices for the Gemini API.

How it works

It provides a library of "skills"—lightweight context packages—that can be added to AI coding assistants. These skills provide the model with the most up-to-date information on how to interact with Gemini models and APIs, improving the accuracy of the code it generates.

Who it’s for

Developers building applications powered by the Gemini API, specifically those using AI coding assistants to accelerate their development workflow.

Highlights

  • Comprehensive API Coverage: Includes skills for the standard Interactions API (text, chat, function calling, structured output), the Live API (real-time bidirectional streaming), and specialized video generation with Gemini Omni 1.1 Flash.
  • Proven Performance: Evaluations show that adding these skills improves an agent's ability to generate correct API code to 87% with Gemini 3 Flash and 96% with Gemini 3.1 Pro.
  • Broad Integration: Compatible with various AI coding tools including Vercel skills CLI, Context7, Antigravity, Claude Code, Cursor, and OpenAI Codex.

Related

  • Project
    google/agents-cli

    A CLI and skill set for building, evaluating, and deploying enterprise-grade AI agents on the Gemini Enterprise Agent Platform using Google Cloud.

  • Project
    google-gemini/gemini-live-api-examples

    A collection of examples for the Gemini Live API, enabling developers to build low-latency, real-time voice and video AI agents with multimodal input.

  • Dispatch
    Gemini 3.7 Flash release notes / what's new

    Google DeepMind has released Gemini 3.7 Flash, a model optimized for coding and agents that offers significant performance gains over Gemini 3.6 Flash at half the introductory cost.

  • Project
    googleapis/js-genai

    Official Google npm SDK (`@google/genai`) that wraps Gemini’s public and Enterprise Agent APIs for JavaScript/TypeScript. Provides simple classes for one‑shot generation, streaming, stateful chats, file uploads, live sessions, and a unified Interactions API. Supports function/tool calling, automatic continuation, and multimodal inputs. Install with `npm i @google/genai`; use an API key or enterprise credentials. Documentation and samples are hosted on the repo’s website.

  • Project
    googleapis/python-aiplatform

    The Python SDK for the Gemini Enterprise Agent Platform (formerly Vertex AI) allows developers to build, deploy, and optimize AI agents, manage prompts, and evaluate generative AI models.