From 2f682eb020042f8a7c77ad61333836b624c495ab Mon Sep 17 00:00:00 2001
From: Mendon Kissling <59585235+mendonk@users.noreply.github.com>
Date: Fri, 29 May 2026 10:28:21 -0400
Subject: [PATCH] docs: agent improvements (#13269)

* docs: add structured output for agents

* docs: add structured output note

* docs: clarify both agent outputs and add release notes

* docs: when langflow sends events to the playground and chat history
---
 docs/docs/Agents/agents.mdx                | 23 +++++++++++++++++++---
 docs/docs/Components/structured-output.mdx |  5 +++++
 docs/docs/Support/release-notes.mdx        | 13 ++++++++++++
 3 files changed, 38 insertions(+), 3 deletions(-)

diff --git a/docs/docs/Agents/agents.mdx b/docs/docs/Agents/agents.mdx
index c3c8b5ff01..1fd5709e38 100644
--- a/docs/docs/Agents/agents.mdx
+++ b/docs/docs/Agents/agents.mdx
@@ -130,6 +130,10 @@ By default, the **Agent** component uses your Langflow installation's storage, a
 The **Message History** component isn't required for default chat memory, but it is required if you want to use external chat memory like Mem0.
 Additionally, the **Message History** component provides more options for sorting, filtering, and limiting memories. Although, most of these options are built-in to the **Agent** component with default values.
 
+Langflow sends events to the Playground during each agent run: the input message, each tool call with its input and result, streaming tokens as they arrive, and the final answer.
+Langflow writes the completed message to chat history when the run finishes.
+When using the **Structured Response** output, Langflow does not send events to the Playground and does not write to chat history.
+
 For more information, see [Store chat memory](/memory#store-chat-memory) and [**Message History** component](/message-history).
 
 ### Additional parameters
@@ -145,10 +149,23 @@ For example:
 
 ## Agent component output
 
-The **Agent** component outputs a **Response** (`response`) that is [`Message` data](/data-types#message) containing the agent's raw response to the query.
+The **Agent** component has two outputs:
 
-Typically, this is passed to a **Chat Output** component to return the response in a human-readable format.
-It can also be passed to other components if you need to process the response further before, or in addition to, returning it to the user.
+* **Response** (`response`): The agent's reply as [`Message`](/data-types#message) data, which is typically connected to a **Chat Output** component.
+* **Structured Response** (`structured_response`): The agent's reply formatted as structured [`Data`](/data-types#data) according to the **Output Schema** you define.
+
+    To configure the agent's **Structured Response** output:
+
+    1. In the **Agent** component, click the output label near the component's output port and select **Structured Response**.
+    2. Select the **Agent** component to open the [component inspection panel](/concepts-components#component-inspection-panel), and then click <Icon name="Table" aria-hidden="true"/> **Open table**.
+    3. Click <Icon name="Plus" aria-hidden="true"/> **+** to add a row for each field you want to extract. For each row, enter a **Name**, **Type**, and optionally a **Description** and **As List** toggle.
+    4. Connect the **Structured Response** port to a downstream component that accepts [`Data`](/data-types#data) input, such as a **Parser** or **JSON Operations** component.
+
+    The **Agent** component uses the connected LLM to extract structured data, and the extraction behavior is configured in the **Output Format Instructions** field.
+    Modify the prompt in the **Output Format Instructions** field to change extraction behavior.
+    This does not modify the schema defined in **Output Schema**.
+
+    Both outputs can be connected at the same time, but each output produces a separate LLM call. If you only need structured data, connect **Structured Response** only.
 
 ## See also
 
diff --git a/docs/docs/Components/structured-output.mdx b/docs/docs/Components/structured-output.mdx
index 4bc8403705..efb734934d 100644
--- a/docs/docs/Components/structured-output.mdx
+++ b/docs/docs/Components/structured-output.mdx
@@ -14,6 +14,11 @@ For example, you can extract specific details from documents, like email message
 
 ## Use the Structured Output component in a flow
 
+:::tip
+If you're building an agentic flow, the **Agent** component includes a built-in **Structured Response** output that produces structured `Data` without a separate **Structured Output** component.
+For more information, see [Agent component output](/agents#agent-component-output).
+:::
+
 To use the **Structured Output** component in a flow, do the following:
 
 1. Provide an **Input Message**, which is the source material from which you want to extract structured data.
diff --git a/docs/docs/Support/release-notes.mdx b/docs/docs/Support/release-notes.mdx
index fddc58af0f..645787b8b9 100644
--- a/docs/docs/Support/release-notes.mdx
+++ b/docs/docs/Support/release-notes.mdx
@@ -69,6 +69,19 @@ For all changes, see the [Changelog](https://github.com/langflow-ai/langflow/rel
 
 ### New features and enhancements
 
+- **Agent** component: structured response output
+
+    The **Agent** component now includes a **Structured Response** (`structured_response`) output that returns the agent's reply as structured data according to an **Output Schema** you define.
+
+    For more information, see [Agent component output](/agents#agent-component-output).
+
+- **Agent** component: default system prompt
+
+    The **Agent** component now uses a structured default system prompt that includes the current date and the active model name.
+    Existing flows are not affected.
+
+    For more information, see [Agent instructions and input](/agents#agent-instructions-and-input).
+    
 - Extension bundles
 
     Langflow 1.10 introduces Extension bundles: component providers that are packaged and versioned as standalone pip packages, independent of the core Langflow server.