Skip to main content

Quick Overview

This product walkthrough from Skill Leap AI introduces the Skills feature recently added to Google Gemini. The presenter demonstrates how to access, configure, and execute custom task instructions across both Gemini Chat and the Spark AI agent. The video serves as a quick guide for streamlining repetitive document processing workflows.

Key Points

  • 1.Google Gemini now includes a Skills feature that works within standard Gemini Chat and the Spark AI agent interface.
  • 2.Users can create custom skills either manually or automatically through Gemini using natural language prompts.
  • 3.Skills store specific formatting rules and task instructions so users do not have to repeat context or prompting steps.
  • 4.In a new chat session, users can trigger an active skill by typing a forward slash command.
  • 5.Supplying a source file such as a meeting transcript to an invoked skill automatically generates standardized structured output.

Summary

The video begins with an overview of a new feature in Google Gemini known as Skills. The presenter explains that inside the Gemini web interface, users can toggle between Gemini Chat and Spark, which functions as an AI agent that performs tasks. The Skills section is located in the left navigation menu and integrates with both Spark and the standard Gemini Chat environment.

The interface offers two primary methods for establishing skills, allowing users to build them manually or create them automatically with Gemini. In the demonstration, the presenter initiates skill creation by typing a prompt starting with create a skill to configure a tool called meeting_recap. This skill is designed to take raw meeting notes or full transcripts and structure them into three distinct sections: decisions made, action items with owners and due dates, and open questions.

After Gemini generates the configuration, the interface confirms that the meeting recap skill is saved and active. The screen displays the generated system instructions, detailing the exact triggers, extraction guidelines, and output formatting rules that the skill will follow whenever it is executed.

To use the newly built skill, the presenter opens a new chat window and types a forward slash into the prompt bar to bring up the active skill list. Selecting the meeting recap skill inserts the command directly into the text box. The presenter then uploads a meeting transcript file without adding extra explanatory prompt text or background context. Gemini processes the document and outputs a formatted breakdown matching the predetermined template, eliminating the need to retype context or instructions.

Introduction to Gemini Skills and Spark

Google Gemini has introduced a Skills feature accessible through both the standard Chat interface and Spark, an AI agent designed to perform tasks. The Skills tab allows users to configure automated workflows and reusable custom instructions.

Creating a Custom Skill

Skills can be set up manually or generated automatically by prompting Gemini with a command such as create a skill. In the demonstration, a skill is created to process meeting notes and transcripts into standardized recaps containing decisions made, action items, and open questions.

Executing Skills with Slash Commands

Once saved and marked active, a skill can be invoked in any new chat by typing a forward slash into the input bar. Users can attach a transcript or document file directly to the skill command, producing consistent, structured output without retyping prompts or context.

The Bottom Line

The demonstration establishes how the Skills update in Google Gemini allows users to create reusable instruction sets for recurring tasks like meeting recaps. It confirms that skills can be triggered with a simple slash command in Gemini Chat or Spark alongside uploaded documents. The video focuses exclusively on single-task creation and slash invocation, leaving advanced multi-step automations or broader application integrations unexplored.

FAQ

What is Google Gemini Skills and how does it function inside the interface?

Google Gemini Skills is a feature that allows users to create and store reusable instructions for tasks, operating inside standard Gemini Chat as well as the Spark agent interface.

What is the Spark AI agent inside the Google Gemini web interface?

Spark is an AI agent within Google Gemini designed to perform tasks and execute workflows on behalf of the user.

How can users create a new custom skill in Google Gemini?

Users can create a skill manually in the settings menu or generate one automatically by entering a prompt starting with create a skill into Gemini.

How do users invoke an active skill inside a new Google Gemini chat?

Users can invoke a saved skill by typing a forward slash into the chat box, selecting the skill name, and uploading the relevant file or text.

What sections were included in the demonstrated Google Gemini meeting recap skill?

The meeting recap skill was configured to format summaries into three sections: decisions made, action items with assigned owners and due dates, and open questions.

Worth watching for

Professionals and regular Google Gemini users who want to automate repetitive formatting tasks, document summaries, and meeting notes without writing repetitive prompts.

  • google-gemini
  • gemini-skills
  • ai-agents
  • spark
  • prompt-engineering
  • meeting-recap