Gemini Live API

Headlines · 2

1. Nov 6
   Gemini API Agent Skills Repository Provides Installable Development Skills

   The google-gemini/gemini-skills repository provides installable skills for building applications with the Gemini API, SDK and model interactions. It covers areas including the Interactions API, Gemini Live API, CLI workflows and Gemini Omni 1.2 Flash, and can be installed through coding assistants such as Claude Code, Cursor and OpenAI Codex. The repository reports that adding the skills improved an agent’s ability to generate correct API code following best practices to 87% with Gemini 3 Flash and 96% with Gemini 3.1 Pro. Those evaluation results have not been independently verified, and the project is not an officially supported Google product.

   GitHub Trending · Python · 🔥 0
2. Oct 19
   Google ADK guide: keeping voice agents conversational during slow tool calls

   Google Developers Blog published a step-by-step guide by Stephen Allen and Kaz Sato on building voice agents that stay conversational while slow tool calls run, using the open-source Agent Development Kit (ADK) and Gemini 3.8 Live. The guide says backend queries cannot always be sub-second, so the fix is architectural: streaming tools, non-blocking FunctionTool definitions, and response scheduling let the agent acknowledge a request immediately, narrate progress, and answer unrelated questions while a query continues in the background. The authors report that in their own sales-report benchmark on Gemini 3.8 Live, switching from a blocking tool to a streaming tool cut time to the agent's first words from 14.4 seconds to 1.2 seconds. The post also says non-blocking tools do not shorten the backend query itself. Requirements listed are google-adk 2.10.0 or later, Gemini 3.8 Live on Gemini Enterprise, and bidirectional live sessions (run_live) for streaming tools and live scheduling.

   Google Developers Blog · 🔥 0

Experience and discussion from the community

Share my Gemini Live API experienceAsk about Gemini Live API