AWS sagemaker-unified-studio documentation change
Summary
Restructured documentation to focus on chat playground usage, added detailed step-by-step instructions for interacting with models, removed some chat agent app implementation details, and added image-based prompt examples
Security assessment
Changes focus on user workflow improvements and feature explanations rather than addressing security vulnerabilities. The added 'contact your administrator' note is routine access control guidance, not evidence of a security fix. No new security features are documented beyond existing authentication references.
Diff
diff --git a/sagemaker-unified-studio/latest/userguide/getting-started-use-amazon-bedrock-ide.md b/sagemaker-unified-studio/latest/userguide/getting-started-use-amazon-bedrock-ide.md index 9f47c541c..f00373047 100644 --- a//sagemaker-unified-studio/latest/userguide/getting-started-use-amazon-bedrock-ide.md +++ b//sagemaker-unified-studio/latest/userguide/getting-started-use-amazon-bedrock-ide.md @@ -4,0 +5,2 @@ +Chat with a model in the chat playground + @@ -7 +9 @@ -Get started with Amazon Bedrock in SageMaker Unified Studio by experimenting with a model in a [playground](./bedrock-playgrounds.html) or by creating a [chat agent app](./create-chat-app.html). +Get started with Amazon Bedrock in SageMaker Unified Studio by experimenting with a model in a [playground](./bedrock-playgrounds.html). @@ -11 +13 @@ The Amazon Bedrock in SageMaker Unified Studio playgrounds that lets you easily -A chat agent app allows users to chat with an Amazon Bedrock model through a conversational interface, typically by sending prompts (text or image) and receiving responses. You can integrate the following Amazon Bedrock capabilities into a chat agent app: +After trying the chat and image playgrounds, you can try creating a chat agent app or flows app. A chat agent app allows users to chat with an Amazon Bedrock model through a conversational interface, typically by sending prompts (text or image) and receiving responses. To create a chat agent app, see [Build a chat agent app with Amazon Bedrock in SageMaker Unified Studio](./create-chat-app.html). @@ -13 +15 @@ A chat agent app allows users to chat with an Amazon Bedrock model through a con - * **[Data sources](./data-sources.html)** — Enrich model responses by including context generated from an Amazon Bedrock knowledge base or a single file. +You can also create a [flows app](./create-flows-app.html) that lets you visually design the flow of an app. @@ -15 +17 @@ A chat agent app allows users to chat with an Amazon Bedrock model through a con - * **[Guardrails](./guardrails.html)** — Lets you implement safeguards for your chat agent app based on your use cases and responsible AI policies. +## Chat with a model in the chat playground @@ -17 +19 @@ A chat agent app allows users to chat with an Amazon Bedrock model through a con - * **[Functions](./functions.html)** — Lets a model call a function to access a specific capability when handling a prompt. +In these instructions, you use the Amazon Bedrock in SageMaker Unified Studio chat playground to chat with an Amazon Bedrock in SageMaker Unified Studio model. You chat by sending a prompt to the model and answering the response that the model generates. For more information, see [Experiment with the Amazon Bedrock in SageMaker Unified Studio playgrounds](./bedrock-playgrounds.html). @@ -18,0 +21 @@ A chat agent app allows users to chat with an Amazon Bedrock model through a con +If you don't have access to a model, contact your administrator. @@ -19,0 +23 @@ A chat agent app allows users to chat with an Amazon Bedrock model through a con +###### To chat with a model @@ -20,0 +25 @@ A chat agent app allows users to chat with an Amazon Bedrock model through a con + 1. Navigate to the Amazon SageMaker Unified Studio landing page by using the URL from your admininstrator. @@ -22 +27,27 @@ A chat agent app allows users to chat with an Amazon Bedrock model through a con -You can also create a [flows app](./create-flows-app.html) that lets you visually design the flow of an app. + 2. Access Amazon SageMaker Unified Studio using your IAM or single sign-on (SSO) credentials. For more information, see [Access Amazon SageMaker Unified Studio](./getting-started-access-the-portal.html). + + 3. At the top of the page, choose the **Discover**. + + 4. In the **Generative AI** section, choose **Chat playground** to open the chat playground. + + + + 5. In **Type** select **Model** and then select a model to use in **Model**. If you don't see a model, contact your administrator. + + 6. In the **Enter prompt** text box, enter `What is Avebury stone circle?`. + + 7. (Optional) If the model you chose is a reasoning model, you can choose **Reason** to have the model include its reasoning in the reponse. For more information, see [Enhance model responses with model reasoning](https://docs.aws.amazon.com/bedrock/latest/userguide/inference-reasoning.html) in the _Amazon Bedrock user guide_. + + 8. Press Enter on your keyboard, or choose the run button, to send the prompt to the model. Amazon Bedrock in SageMaker Unified Studio shows the response from the model in the playground. + + + + 9. Continue the chat by entering the prompt `Is there a museum there?` and pressing Enter. + +The response shows how the model uses the previous prompt as context for generating its next response. + + 10. Choose **Reset** to start a new chat with the model. + + 11. Influence the model response by doing the following: + + 1. Enter and run a prompt. Note the response from the model. @@ -24 +55 @@ You can also create a [flows app](./create-flows-app.html) that lets you visuall -These getting started instructions show you how to chat with a model in the chat playground. They also show you how to build a chat agent app for a radio station that creates playlists and uses a data source to get upcoming show information. + 2. Choose the configurations menu to open the **Configurations** pane. @@ -26 +57 @@ These getting started instructions show you how to chat with a model in the chat -After completing this section, continue using Amazon Bedrock in SageMaker Unified Studio by going to [Amazon Bedrock in SageMaker Unified Studio](./bedrock.html). + @@ -28 +59 @@ After completing this section, continue using Amazon Bedrock in SageMaker Unifie -###### Topics + 3. Influence the model response by making [inference parameters](./explore-prompts.html#inference-parameters) changes. @@ -30 +61 @@ After completing this section, continue using Amazon Bedrock in SageMaker Unifie - * [Chat with a model in the chat playground](./getting-started-use-amazon-bedrock-ide-playground.html) + 4. Run the prompt again and compare the response with the previous response. @@ -32 +63,31 @@ After completing this section, continue using Amazon Bedrock in SageMaker Unifie - * [Create a radio station app](./getting-started-use-amazon-bedrock-ide-build-chat-app.html) + 12. Choose **Reset** to start a new chat with the model. + + 13. 14. Try sending an image to a model by doing the following: + + 1. For **Model** , choose a model that supports [images](https://docs.aws.amazon.com/bedrock/latest/userguide/models-supported.html). + + 2. Choose the attachment button at the left of the **Enter prompt** text box. + + + + 3. In the open file dialog box, choose an image from your local computer. + + 4. In the text box, next to the image that you uploaded, enter `What's in this image?`. + + 5. Press Enter on your keyboard enter to send the prompt to the model. The response from the models describes the model or image. + + 15. (Optional) Try using another model and different prompts. Different models have different recommendations for creating, or engineering, prompts. For more information, see [Prompt engineering guides](./explore-prompts.html#prompt-guides). + + 16. (Optional) Compare the output from multiple models, or [shared apps](./bedrock-explore-chat-playground-app.html). + + 1. In the playground, turn on **Compare mode**. + + 2. In both panes, select the model that you want to compare. If you want to use a shared app, select **App** in **Type** and then select the app in **App**. + + 3. Enter a prompt in the text box and run the prompt. The output from each model is shown. You can choose the copy icon to copy the prompt or model response to the clipboard. + + 4. (Optional) Choose **View configs** to make configuration changes, such as [inference parameters](./explore-prompts.html#inference-parameters). Choose **View chats** to return to the chat page. + + 5. (Optional) Choose **Add chat window** to add a third window. You can compare up to 3 models or apps. + + 6. Turn off **Compare mode** to stop comparing models. @@ -45 +106 @@ Create a project -Chat with a model in the chat playground +Get started with the query editor