5.1. AI Coding Assistants¶
AI-powered coding assistants are rapidly becoming indispensable for developers - providing suggestions and automated features that help in writing code more efficiently and accurately. As popularity for these features continues to increase, so does the demand for development environments that can provide such functionality. AI code assist functionality is available with Code Composer Studio using functionality bundled with the IDE, as well as numerous third-party AI code assist extensions available on Open VSX. Code Composer Studio also offers MCP (Model Context Protocol) servers that enable AI assistants to interact directly with the Code Composer Studio project and debug servers, allowing for more integrated and context-aware development workflows.
Important
AI coding assistants are designed to assist developers by providing suggestions and automating certain tasks. However, they are not perfect and may occasionally provide incorrect or suboptimal suggestions. It is important for developers to critically evaluate the suggestions provided by AI coding assistants and use their own judgment when deciding whether to accept or modify those suggestions.
5.2. AI Assistant Configuration¶
The AI Assistant Configuration view allows you to configure and manage supported AI coding assistants. You can access this view by using the Set up an AI assistant link from the Get Started view or by navigating to Help > Get Started with AI.
The AI Assistant Configuration view appears as below:
5.2.1. Step 1: Select agent¶
The first step in configuring an AI coding assistant is to select the agent you want to use. Currently, only Claude Code is supported, but more agents will be added in the future.
If the Claude Code extension is not already installed, it will be installed automatically. Please ensure the proxy is configured correctly in Code Composer Studio to allow for the installation of the extension.
5.2.2. Step 2: Configure model¶
After selecting the agent, you can configure the model settings. If you have an existing subscription with the provider, or already have already configured your model in the past, you can enable the checkbox option Use my existing model configuration and proceed to the next step. If you do not have an existing subscription or want to configure a new model with an API key, you can provide the necessary details below:
Provider: Select the provider for the AI model you want to use.
API Key: Enter the API key for the selected provider to authenticate your access to the AI model.
Base URL: Provide the base URL for the API endpoint of the selected provider.
Model: Choose the specific AI model you want to use for code assistance.
Note
An existing API key must be available for the provider you select in order to manually configure the assistant with the AI Assistant Configuration view. Please refer to the provider's documentation for instructions on how to obtain an API key.
5.2.3. Step 3: Enable MCP servers¶
To allow the AI assistant to interact with Code Composer Studio functionality, you need to enable the MCP (Model Context Protocol) servers. This can be done by checking the Enable CCStudio ecosystem MCP servers option in the configuration view. Enabling this option allows the AI assistant to access Code Composer Studio functionality such as building, debugging, monitoring, and controlling SysConfig.
MCP servers are only active when Code Composer Studio is running and a workspace configured for AI functionality is active. When enabled, the configuration will attempt to start the MCP servers automatically. The servers are stopped when Code Composer Studio is closed or when the option is disabled.
Once all the steps are complete, press the Apply button to save your configuration. If the Claude Code extension is not already installed, it will be installed automatically. After installation, the AI coding assistant will be configured for use within Code Composer Studio with the current workspace. If using Claude Code, open Claude Code Chat with the Open Claude Code link or with the menu option View > Claude Code Chat to start using the assistant.
Note
Only the Claude Code extension is installed by Code Composer Studio. The Claude Code desktop and CLI terminal is not installed as part of this process. If you wish to use the desktop version of Claude Code, it will need to be installed separately. Note that some features of the Claude Code extension may require the desktop version to be installed and running.
Warning
Code Composer Studio will install the latest available version of the Claude Code extension. Due to the frequent updates to the extension, there may be occasional issues with new versions. If you encounter any issues after installation, please refer to the FAQ for troubleshooting steps and potential workarounds.
Note
The AI Assistant Configuration settings are typically stored on a per workspace basis. This allows you to have different AI assistant configurations for different projects or workspaces in Code Composer Studio. When you switch between workspaces, the AI assistant configuration will be loaded based on the settings saved for that specific workspace. When using a new workspace, you will need to configure the AI assistant settings for that workspace before using the assistant features.
5.3. Using Claude Code¶
Important
Claude Code requires Git to manage the local repository that it uses to store conversation history and other relevant data. Please ensure Git is installed and properly configured on your system before using Claude Code.
Once the Claude Code extension is installed and configured, you can start using it to assist with your coding tasks. To open the Claude Code chat interface, use the Open Claude Code link in the AI Assistant Configuration view or navigate to View > Claude Code Chat. This will open the chat interface where you can interact with the Claude Code assistant.
If a subscription will be used and the option to Use my existing model configuration was enabled, the below login prompt will appear in the chat interface:
Select the relevant option to log in to your account and authenticate the connection between the AI assistant and your subscription. Once authenticated, the AI assistant will be able to access the configured model and provide code assistance based on that model.
Once the AI assistant is set up and authenticated, you can start using it to assist with your coding tasks. You can ask questions, request code suggestions, or seek help with debugging and other coding tasks. The assistant will provide responses based on the context of your project and the information you provide in the chat. You can also use the chat interface to manage your conversation history and access other features provided by the Claude Code extension.
For more information on using the Claude Code extension and its features, please refer to the Use Claude Code in VS Code documentation.
5.4. Using other AI Assistants¶
While Claude Code is the only AI assistant fully supported for use within Code Composer Studio, it is possible to use other AI assistants that are available as extensions on Open VSX. To use a different AI assistant, you will need to install the relevant extension from Open VSX and configure it according to the instructions provided by the extension developer. Once the extension is installed and configured, you can use it within Code Composer Studio to assist with your coding tasks. Please refer to the documentation for the specific AI assistant extension you are using for more information.
Please refer to the Using AI Code Assistants with Code Composer Studio documentation more details on some extentions tested with Code Composer Studio and instructions on how to get started with them.
Note that the Code Composer Studio MCP servers are designed to work with any AI assistant that supports the Model Context Protocol, so you may be able to use the MCP servers with other AI assistants as well. However, the level of integration and functionality may vary depending on the specific AI assistant and its capabilities. Claude Code is currently the only AI assistant that has been fully tested and verified to work with the Code Composer Studio MCP servers, so if you choose to use a different AI assistant, you may need to do some additional configuration and testing to ensure it works properly with Code Composer Studio.
To configure the MCP servers to work with a different AI assistant, you will need to refer to the documentation for that specific assistant to understand how to set up the connection and ensure compatibility with the Model Context Protocol. The below information provides the command and configuration file information for the MCP servers in Code Composer Studio, which will be required when configuring other AI assistants to work with Code Composer Studio:
Command:
[CCS INSTALL DIR]/ccs/tools/node/nodeArguments:
Debug MCP Server:
[USER APPLICATION DATA PATH]/Texas Instruments/CCS/[CCS INSTALL DIR]/0/theia/deployedPlugins/ccs-ai-[VERSION]/extension/dist/mcp-server-proxy.js debugBuild MCP Server:
[USER APPLICATION DATA PATH]/Texas Instruments/CCS/[CCS INSTALL DIR]/0/theia/deployedPlugins/ccs-ai-[VERSION]/extension/dist/mcp-server-proxy.js projectSysConfig MCP Server:
[USER APPLICATION DATA PATH]/Texas Instruments/CCS/[CCS INSTALL DIR]/0/theia/deployedPlugins/ccs-ai-[VERSION]/extension/dist/mcp-server-proxy.js sysconfig
[CCS INSTALL DIR] refers to the directory where Code Composer Studio is installed on your system.
[USER APPLICATION DATA PATH] refers to the path where user application data is stored on your system, which can vary depending on the operating system you are using.
[VERSION] refers to the version of the ccs-ai extension that is installed in your Code Composer Studio environment.
5.5. Theia AI¶
Theia AI is an extension to the Theia IDE framework that integrates artificial intelligence (AI) capabilities to enhance the software development experience within the IDE. It comes bundled with Code Composer Studio and can be used as an AI coding assistant option. Please refer to the Theia AI section of the Using AI Code Assistants with Code Composer Studio documentation.