The "Career Coach" Declarative Agent is an open-source AI-driven tool designed to assist professionals in their career development. This agent provides personalized suggestions and actionable plans to help users learn and grow in their careers. By leveraging data such as the user's current role, skills, and career aspirations, the Career Coach offers tailored advice on skill development, learning opportunities, and career transitions. It maintains a professional and supportive tone, ensuring that interactions are contextual and relevant. The agent integrates with OneDrive, SharePoint, and Graph Connectors to enhance its capabilities. Key features include creating detailed career development plans, performing skill gap analyses, recommending learning resources, and offering networking strategies. This open-source project aims to empower individuals to achieve their career goals through structured guidance and support.
With the declarative agent, you can build a custom version of Copilot that can be used for specific scenarios, such as for specialized knowledge, implementing specific processes, or simply to save time by reusing a set of AI prompts. For example, a grocery shopping Copilot declarative agent can be used to create a grocery list based on a meal plan that you send to Copilot.
Prerequisites
To run this app template in your local dev machine, you will need:
- Node.js, supported versions: 16, 18
- A Microsoft 365 account for development.
- Teams Toolkit Visual Studio Code Extension version 5.0.0 and higher or Teams Toolkit CLI
- Microsoft 365 Copilot license
- First, select the Teams Toolkit icon on the left in the VS Code toolbar.
- In the Account section, sign in with your Microsoft 365 account if you haven't already.
- Create Teams app by clicking
Provisionin "Lifecycle" section. - Select
Preview in Copilot (Edge)orPreview in Copilot (Chrome)from the launch configuration dropdown. - Once the Copilot app is loaded in the browser, click on the "…" menu and select "Copilot chats". You will see your declarative agent on the right rail. Clicking on it will change the experience to showcase the logo and name of your declarative agent.
- Ask a question to your declarative agent and it should respond based on the instructions provided.
| Folder | Contents |
|---|---|
.vscode |
VSCode files for debugging |
appPackage |
Templates for the Teams application manifest, the GPT manifest and the API specification |
env |
Environment files |
The following files can be customized and demonstrate an example implementation to get you started.
| File | Contents |
|---|---|
appPackage/declarativeAgent.json |
Define the behaviour and configurations of the declarative agent. |
appPackage/instruction.txt |
Specific instruction for the declarative agent. In a TXT file for ease of edition automatically scafold by TTK when solution is provisionned. |
appPackage/manifest.json |
Teams application manifest that defines metadata for your declarative agent. |
The following are Teams Toolkit specific project files. You can visit a complete guide on Github to understand how Teams Toolkit works.
| File | Contents |
|---|---|
teamsapp.yml |
This is the main Teams Toolkit project file. The project file defines two primary things: Properties and configuration Stage definitions. |

