Skip to main content

Key Concepts

This section contains the main concepts related to StackSpot AI, which will help you understand how the platform works and how to use it efficiently.

Agents

Agents are customizable entities that use contextual information to provide more precise and relevant responses and actions. They are designed to solve specific problems and improve software development efficiency.

Agent Features

  • Customizable: You can configure Agents to meet the specific needs of your project.
  • Context-based: They use information from Knowledge Sources and Quick Commands to enrich their responses.
  • Flexible: They can be used for various tasks, such as:
    • Generating user stories;
    • Code review;
    • Identifying security vulnerabilities;
    • Creating documentation.

Depending on the objective, agents can be configured as Conversational Agents or Systematic Agents.

  • Conversation Agents: For ongoing interactions that maintain context and allow multiple exchanges.

  • Systematic Agents: For quick, direct responses that do not rely on previous context.

  • Learn how to create Agents;

Knowledge Sources

Knowledge Sources are a crucial feature of StackSpot AI that enables you to create and organize contextualized and personalized information, helping to form a comprehensive knowledge base. This content can include code snippets, API contracts, technical documentation, and architecture patterns. The goal is to enhance the platform's context, making AI responses more accurate, relevant, and aligned with the needs of your team or organization.

What does a Knowledge Source do

When you add a Knowledge Source, you provide context for StackSpot AI. This allows you to supply specific information that will be used to generate more relevant responses and code suggestions. StackSpot AI utilizes this data to identify similarities and references, ensuring that the answers align with your context's practices, standards, and integrations.

Why use Knowledge Sources?

  • Centralized Knowledge: It consolidates important information in one place, making it easier to access and share among teams.

  • Standardization: It ensures that everyone uses the same references, reducing errors and inconsistencies.

  • Increased Productivity: It automates responses and suggestions based on the real context of your project.

  • Additional Information: Knowledge Sources can include code snippets, APIs, documentation, and more.

Usage example

If your team has several internal APIs, you can add their contracts as Knowledge Sources. This will enable StackSpot AI to suggest code implementations that are already integrated with these APIs, saving time and ensuring compliance with internal standards.

Other examples:

  • Adding development best practices;

  • Creating a knowledge base for security policies;

  • Providing technical documentation for reference.

  • Learn more about Knowledge Sources.

Quick Commands

Quick Commands are predefined instructions that can be sent to the Large Language Model (LLM) to perform specific actions. They offer an efficient way to contextualize the LLM and automate repetitive tasks.

Features of Quick Commands

  • User-friendly: Quick Commands can be easily accessed in the IDE by right-clicking or typing / in the chat.
  • Flexible: They can be created in the StackSpot AI Portal and executed in the IDE or through the API.
  • Usage examples:
    • Translate text from Portuguese to English.
    • Search for security violations in the code based on company policies.
    • Add content to the organization’s Workspace.

Quick Commands are a powerful tool designed to enhance productivity and ensure consistency in development.

Stack AI

Stack AI is a declarative technology set that enhances the context available to the Large Language Model (LLM). It enables the generation of more sophisticated and refined code aligned with the project's needs.

How does Stack AI work?

  • Specific definitions: Includes information such as:

    • Programming language;
    • Frameworks;
    • Design patterns;
    • Cloud attributes and providers.

  • Advanced contextualization: By specifying Stack AI technologies within a Spots, you provide inputs that allow the system to generate more precise and project-aligned code suggestions.

  • Create your StackSpot AI.

Spot

A Spot is a collaborative space within StackSpot AI designed to organize, share, and manage content such as Agents, Knowledge Sources (KS), Stacks AI, and Quick Commands with specific users or teams. Spots bring together content and members, allowing you to control who can view or edit the shared resources.

Benefits of using Spots

  1. Centralized collaboration: Collect related resources and team members in one location, making managing and accessing shared knowledge easier.
  2. Permission control: Determine who can view (Read) or modify (Write) content, ensuring appropriate access levels for each member.
  3. Content independence: Adding personal content to a Spot creates an independent copy. Changes made within a Spot do not affect the original content, providing flexibility.
  4. Automatic updates: Updates made at the Account level are automatically reflected in all associated Spots, ensuring that shared content remains current.
  5. Streamlined sharing: Quickly share curated tools and knowledge with your team, enhancing consistency and productivity.

Next Step