AWS bedrock documentation change
Summary
Updated documentation for built-in tools in specific OpenAI models, simplified tool descriptions, removed parameter tables, and added code examples for tool usage.
Security assessment
The changes are purely functional/documentation updates about tool usage in specific OpenAI models. There is no mention of security vulnerabilities, patches, or security-related configurations. Changes focus on tool behavior and integration without addressing security risks or security features.
Diff
diff --git a/bedrock/latest/userguide/tool-use.md b/bedrock/latest/userguide/tool-use.md index 245e55713..f82e81644 100644 --- a//bedrock/latest/userguide/tool-use.md +++ b//bedrock/latest/userguide/tool-use.md @@ -462 +462 @@ When you specify a Lambda function in the `tools` parameter, the API will attemp -There are two AWS provided tools that are available in Bedrock: Note-taking functionality (notes tool) and Task management (tasks tool). Let us go over them both in details. +There are two AWS provided tools built into the `openai.gpt-oss-20b` and `openai.gpt-oss-120b` models: Note-taking functionality (notes tool) and Task management (tasks tool). These tools are automatically available — you do not need to define them in the `tools` parameter. @@ -466 +466 @@ There are two AWS provided tools that are available in Bedrock: Note-taking func -The `notes` tool allows you to store and retrieve key-value pairs within the same conversation session. This provides a simple memory mechanism for maintaining context across multiple interactions. Keys are case-sensitive strings and there is no limit on key length or naming conventions. The system overwrites previous values for the same key. Values are stored as strings (JSON, URLs etc.) and no size limits enforced at tool level. Values are persisted through the entire conversation session. Memory is scoped to the current conversation only. +The `notes` tool allows the model to store notes within the same conversation session. This provides a simple memory mechanism for maintaining context across multiple interactions. Memory is scoped to the current conversation only. @@ -468 +468 @@ The `notes` tool allows you to store and retrieve key-value pairs within the sam -**Parameters** +When the model uses the notes tool, it emits an `mcp_call` output with `name` set to `"notes"`. The model determines the appropriate arguments based on your request. @@ -470,5 +470 @@ The `notes` tool allows you to store and retrieve key-value pairs within the sam -Parameter | Type | Required | Description ----|---|---|--- -`operation` | string | Yes | The operation to perform: `"store"` or `"recall"` -`key` | string | Yes | The key identifier for the memory item -`value` | string | Conditional | The value to store (required only for `"store"` operation) +You can use either natural language (e.g. "Remember that my favorite color is blue", "What did I tell you about my favorite color?", "Store the fact that I prefer morning meetings", "Recall what I said about meeting preferences") or you can use direct tool calls in your prompt ("Use the notes tool to store my email as [email protected]", "Check the notes for my email address"). @@ -476 +472 @@ Parameter | Type | Required | Description -**Valid Operations** +**Tasks Tool Overview** @@ -478 +474 @@ Parameter | Type | Required | Description - * **`store`** : Save a key-value pair to memory +The `tasks` tool provides a stack for managing tasks within a conversation session. You can push tasks onto the stack and pop them off, making it useful for managing workflows, reminders, or hierarchical task management. Tasks persist throughout the entire conversation session. Memory is scoped to the current conversation only. @@ -480 +476 @@ Parameter | Type | Required | Description - * **`recall`** : Retrieve a value by its key +When the model uses the tasks tool, it emits an `mcp_call` output with `name` set to `"tasks"`. The model determines the appropriate arguments (such as `method`, `task.title`, and `task.description`) based on your request. @@ -481,0 +478 @@ Parameter | Type | Required | Description +You can call the Tasks tool by either using natural language (e.g. "Add a task to review the budget", "Push a reminder to call the client", "What's the next task I need to do?", "Pop the most recent task", "Get the latest task from my stack") or you can call the tool directly in your prompt ("Use the tasks tool to push 'finish presentation'", "Pop a task from the stack", "Add 'schedule meeting' to my task list"). @@ -482,0 +480 @@ Parameter | Type | Required | Description +**Code example: Using the notes and tasks tools** @@ -483,0 +482 @@ Parameter | Type | Required | Description +The notes and tasks tools are built into the `openai.gpt-oss-20b` and `openai.gpt-oss-120b` models. You do not need to define them explicitly in the `tools` parameter — simply reference them in your prompt: @@ -485 +483,0 @@ Parameter | Type | Required | Description -You can use either natural language (e.g. "Remember that my favorite color is blue", "What did I tell you about my favorite color?", "Store the fact that I prefer morning meetings", "Recall what I said about meeting preferences") or you can use direct tools calls in your prompt ("Use the notes tool to store my email as [email protected]", "Check the notes for my email address"). @@ -487 +485,5 @@ You can use either natural language (e.g. "Remember that my favorite color is bl -**Tasks Tool Overview** + from openai import OpenAI + + client = OpenAI( + base_url="https://bedrock-mantle.us-east-1.api.aws/v1" + ) @@ -489 +491,8 @@ You can use either natural language (e.g. "Remember that my favorite color is bl -The `tasks` tool provides a Last-In-First-Out (LIFO) stack for managing tasks within a conversation session. This allows you to push tasks onto a stack and pop them off in reverse order, making it perfect for managing nested workflows, temporary reminders, or hierarchical task management. Tasks persist throughout the entire conversation session. Stack state is maintained across multiple interactions. Memory is scoped to the current conversation only. + # The notes tool is built-in — just ask the model to use it + resp = client.responses.create( + model="openai.gpt-oss-120b", + input="Use the notes tool to store that my preferred language is Python.", + ) + + print(resp.output) + # The model automatically calls the notes tool via mcp_call @@ -491,5 +500,5 @@ The `tasks` tool provides a Last-In-First-Out (LIFO) stack for managing tasks wi -Parameter | Type | Required | Description ----|---|---|--- -`operation` | string | Yes | The operation to perform: `"push"` or `"pop"` -`description` | string | Conditional | The description of the task item (required only for `"push"` operation) -`summary` | string | No | Optional summary about the task item (only for `"push"` operation) + # Use the tasks tool to push a task + resp = client.responses.create( + model="openai.gpt-oss-120b", + input="Use the tasks tool to push a task: review the API documentation", + ) @@ -497 +506 @@ Parameter | Type | Required | Description -You can call the Tasks tool by either using natural languages (e.g. "Add a task to review the budget", "Push a reminder to call the client", "What's the next task I need to do?", "Pop the most recent task", "Get the latest task from my stack") or you can call the tool directly in your prompt ("Use the tasks tool to push 'finish presentation'", "Pop a task from the stack", "Add 'schedule meeting' to my task list"). + print(resp.output)