Today, we’re excited to introduce Qodo Gen 1.0 to the community!
Agentic chat
Qodo Gen now supports either Standard or Agentic workflow modes. Standard mode offers a more manual Qodo Gen experience, including Commands.
Agentic Mode makes interactions more dynamic and less structured, proactively assisting with tasks and providing contextual suggestions.
The agentic chat is the evolution to multi-step problem-solving rather than one-shot AI responses. Qodo Gen’s agentic chat is autonomous in that instead of just writing code or answering a question just once, the chat agent can make decisions, ask questions, use tools and carry out tasks autonomously.
The coding agent leverages core Qodo tools for understanding prompt intent and advanced context retrieval that include codebase indexing and analysis, web search and scraping.
Here are some of the things the agent can do:
- Write New Code: Generate a complete REST API using Flask, including routes, models, and database setup.
- Fix Bugs: Help identify and resolve a memory leak in a Java application by analyzing heap dumps and suggesting optimizations.
- Implement a Task from Text or a Ticket: Translate a user story into a working feature in a React application, complete with state management and API integration.
- Develop Frontend Components: Build reusable React components with Tailwind CSS for a responsive web application.
This AI-powered conversational interface accelerates development while keeping you in control.
Extensible Agent & Tooling Support
In order to understand your code better and help you perform various actions, Agentic Mode uses multiple services behind the scenes. In addition to built-in servers, users can add their own external tools via Anthropic’s Model Context Protocol (MCP). Users who have installed the MCP server can bring 3rd party or custom tools and agents to enhance their workflows by adding the API key.
- Pre-configured MCP tools (like Jira and GitHub) can be activated with an API key.
- Custom MCP tools allow you to connect other services by adding them manually in Qodo Gen or using Smithery.
Built-in agentic tools
- Fetch Service: Retrieves web content in various formats.
- File System Service: Manages files and directories.
- Git Service: Provides Git repository insights.
- LSP (Language Server Protocol) Service: Analyzes and tracks code structure.
This extensibility ensures that Qodo Gen 1.0 seamlessly integrates into your development environment, allowing you to enhance workflows with both built-in capabilities and custom tools tailored to your needs.
Semi-agentic test generation
In previous versions of Qodo Gen, testing capabilities were carried out in a panel within the IDE. This advanced testing suite was dedicated to code behavior analysis, test behavior generation and test suite expansion.
In Qodo Gen 1.0, users now have the option to use a chat-based, semi-agentic test generation workflow. Users can follow a step-by-step flow to add context, example tests and mocks, generate test behaviors and comprehensive test suites.
Using Test Generation
- Open a file: Open the file in your project that you want to create tests for.
- Activate Qodo Gen: Click the Qodo Gen logo in the Extensions bar.
- Select Component: Select the component that you’d like to test from the bar underneath the chatbox.
- Start Test Generation: Type /test in the chatbox and hit the send key or click the arrow button. Qodo Gen will start leading you through the Test Generation process.
This new workflow makes test generation more interactive and intuitive, allowing developers to collaborate with Qodo Gen 1.0 in refining, expanding, and optimizing their test suites with greater efficiency.
Get Started with Qodo Gen 1.0
Qodo Gen 1.0 is now available! Whether you're streamlining development, integrating AI-driven automation, or optimizing your test suite, Qodo has the tools to transform your workflow.
Thanks for reading! Questions or comments? Leave them below.
Top comments (4)
Just install on VSCode ✅
amazing, great work team!
Informative, Thank you
The extensible MCP tooling and semi-agentic test generation sound super practical too. Curious how it manages tricky legacy code, but I’m definitely eager to try it out. Congrats on the launch!