How to get started...

  1. Onboarding:

    1. Please request the necessary permission and details for the EYQ API, ensuring you specify the environment. Click here to access the request form

    2. Once your request is processed, the EYQ Team will onboard you to the platform. They will also provide the client details, App scopes and subscription key via email.

  2. Steps to Get Started with EYQ Skill Chat:

    1. Create an Azure AD App Registration:
      After receiving the scopes via email (Step 1), proceed to create an Azure Active Directory (AD) app registration. Make sure to include the provided scopes during this process.

    2. API Request Headers:
      When making API calls, you need to include the following in the request header:

      • x-client-id: Your client ID

      • Ocp-Apim-Subscription-Key: Your subscription key

      • Bearer Access Token: The access token generated from the above app registration.

  3. API Documentation:

    1. Initialize the Conversation:

    Before starting any interaction, you must initialize the conversation by making a `POST` request to the /api/v1/conversations/init endpoint. This is essential as it sets up the session for the chat. To begin a new chat session, you'll need to call this initialization endpoint again, ensuring that a fresh conversation context is created.

    2. Reference Conversation ID for Future Chats:

    After initializing a conversation, the API will return a unique conversation ID. This ID should be used as a reference for all subsequent chat messages within that particular conversation. To maintain continuity and track the flow of the chat, always reference this conversation ID when making API `POST` call to /api/v1/conversations.

    3. Generate a Conversation Title:

    You can generate a title for the conversation by making a `POST` request to the /api/v1/conversations/{conversationId}/generate-title endpoint. This action is necessary only once, and it should be performed when you receive the first response from the conversation API. A well-defined title helps in categorizing and revisiting past conversations efficiently.

    4. Integrate User Feedback:

    If you wish to allow users to provide feedback on the bot’s responses, integrate the /api/v1/conversations/{conversationId}/message-feedback endpoint using a `POST` request. Users can express their satisfaction by liking or disliking a message. Importantly, if a user dislikes a response, providing feedback becomes mandatory, ensuring that constructive criticism is captured to improve future interactions.

    5. Reference Documentation and Downloads:

    In each conversation, the bot may provide references from which the response was generated, along with specific page numbers. If users want to access or download the referenced documents, they can do so by calling the /api/v1/skills/{skillId}/version/{skillVersionId}/document endpoint using a `GET` request. This may include optional query parameters such as `documentName` or `documentId` for precise retrieval.

    6. View Past Conversations:

    Users can review their previous interactions by making a `GET` request to the /api/v1/conversations endpoint. This allows them to filter their search by parameters like `startDate`, `endDate`, `status`, `filters`, `sortOrder`, `sortColumn`, `pageNumber`, and `pageSize`. These parameters provide flexibility in retrieving conversations based on specific criteria and organizing the results in a user-friendly manner.

  4. Traceability:

    To enable tracing requests across the platform, you can generate a unique GUID and include it as an 'x-correlation-id' in the request header. This will help us trace the request if any issues occur.