Setup & Installation
What This Skill Does
Covers building applications with the Gemini API across Python, JavaScript/TypeScript, Go, and Java. Handles text generation, multimodal inputs (images, audio, video, documents), function calling, structured JSON output, code execution, context caching, and embeddings. Includes current model specs and SDK migration guidance.
Instead of manually cross-referencing the REST discovery spec and SDK changelogs to pick the right model and method signatures, this skill provides current model names, correct SDK imports, and working code patterns in one place.
When to use it
- Generating structured JSON responses that match a specific schema from user input
- Calling your own functions from within a Gemini model response
- Processing uploaded video or audio files to extract information
- Building a multi-turn chat interface backed by Gemini models
- Generating text embeddings for semantic search over a document corpus