Cody - Sourcegraph

Description

Cody is an AI-powered coding assistant designed to accelerate the development process by understanding and utilizing your entire codebase. It integrates seamlessly with popular code hosting platforms like GitHub and GitLab, as well as IDEs such as VS Code, JetBrains, and Visual Studio. By connecting to these environments, Cody acts as a personal AI assistant, equipped to enhance code comprehension, generation, and debugging.

At its core, Cody utilizes the latest Large Language Models (LLMs) and draws context from your development environment. This context is pulled using Sourcegraph's advanced Search API, allowing Cody to access information about APIs, symbols, and usage patterns across both local and remote codebases. This comprehensive context ensures that Cody can provide highly accurate and relevant assistance.

Key capabilities of Cody include developer chat, where users can ask questions about their code, generate new code snippets, and request code edits. The assistant offers intelligent code completions and auto-editing suggestions, which analyze cursor movements and recent changes to propose contextual modifications. Cody also features a robust prompt system, enabling developers to automate recurring tasks with customizable and shareable prompts. Furthermore, its debugging capabilities are optimized to identify and fix errors, significantly speeding up the troubleshooting process and boosting developer productivity.

Cody offers flexible deployment options, including extensions for VS Code, JetBrains, and Visual Studio, as well as a web app interface and a command-line interface (CLI). For enterprise-level solutions, Cody for Sourcegraph Enterprise is available. The platform also includes context filters, allowing users to exclude specific repositories from chat and autocomplete results, thereby enhancing control over the context used by the AI.

Data collection by Cody focuses on prompts and responses to provide the service. For individual users on Sourcegraph.com, prompts and responses may be used to enhance user experience but not for model training. Usage data and feedback are also collected to improve the overall user experience. Cody is compatible with other Sourcegraph products, such as Code Search, allowing users to query their codebase and then use Cody's chat interface to ask questions about the search results.

Cody's Core Features

• AI-powered code understanding, writing, and fixing

• Utilizes LLMs and development context for accuracy

• Integrates with GitHub, GitLab, VS Code, JetBrains, Visual Studio

• Developer chat for code-related questions and generation

• Code completions and auto-editing suggestions

• Customizable and shareable prompts for task automation

• Advanced Search API for local and remote codebase context

• Debugging assistance to identify and fix code errors

• Context filters to manage repository inclusion

• Available as IDE extensions, web app, and CLI

• Enterprise solutions for Sourcegraph Enterprise

How to use Cody?

1. 1

   Install: Download and install the Cody extension for your IDE (VS Code, JetBrains, Visual Studio) or access it via the Sourcegraph web app.

2. 2

   Connect: Link Cody to your code repositories (e.g., GitHub, GitLab) and development environment.

3. 3

   Interact: Use the chat interface to ask questions about your code, request code generation, or initiate code edits.

4. 4

   Utilize Completions: Leverage Cody's intelligent code completions as you type.

5. 5

   Configure Prompts: Create and use custom prompts to automate repetitive coding tasks.

6. 6

   Debug: Employ Cody's debugging capabilities to identify and resolve errors in your code.

7. 7

   Manage Context: Configure context filters to specify which repositories Cody should consider.

Cody's Use Cases

• Code Understanding
• Code Generation
• Code Editing
• Debugging Assistance
• Task Automation
• Contextual Search

FAQ from Cody

• What is Cody by Sourcegraph?

  Cody is an AI coding assistant that uses LLMs and your development context to help you understand, write, and fix code faster. It integrates with codehosts and IDEs to provide context-aware assistance.

• How does Cody get context from my codebase?

  Cody uses Sourcegraph's advanced Search API to pull context from both local and remote codebases, including information about APIs, symbols, and usage patterns.

• What IDEs does Cody support?

  Cody is available as extensions for VS Code, JetBrains, and Visual Studio. It can also be used in the Sourcegraph web app.

• Can Cody help with debugging?

  Yes, Cody is optimized to identify and fix errors in your code, accelerating the debugging process and increasing developer productivity.

• How can I automate tasks with Cody?

  Cody allows you to create and use premade or customizable prompts to automate key tasks in your workflow, which can be saved and shared with your team.

• What data does Cody collect?

  Cody collects prompts and responses to provide the service. For individual users on Sourcegraph.com, this data may be used to enhance user experience but not for model training. Usage data and feedback are also collected.

• Is Cody available for enterprise use?

  Yes, Cody is available for Sourcegraph Enterprise. You can get in touch with their team to try it.

• Can I control which repositories Cody uses for context?

  Yes, Cody features context filters that allow you to ignore selected repositories from chat and autocomplete results.

• How does Cody's auto-edit feature work?

  Cody's auto-edit suggests code changes by analyzing cursor movements and typing. After an edit, it proposes contextual modifications based on your cursor position and recent changes.

• Where can I get help or join the community for Cody?

  You can ask questions in their Discord community or create a post on X (formerly Twitter).