Update ADK doc according to issue #744 - 1

docs/agents/llm-agents.md

@@ -160,7 +160,7 @@ on the conversation and its instructions.
       # Replace with actual logic (e.g., API call, database lookup)
       capitals = {"france": "Paris", "japan": "Tokyo", "canada": "Ottawa"}
       return capitals.get(country.lower(), f"Sorry, I don't know the capital of {country}.")
-    
+
     # Add the tool to the agent
     capital_agent = LlmAgent(
         model="gemini-2.0-flash",
@@ -174,7 +174,7 @@ on the conversation and its instructions.
 === "Java"
 
     ```java
-    
+
     // Define a tool function
     // Retrieves the capital city of a given country.
     public static Map<String, Object> getCapitalCity(
@@ -185,13 +185,13 @@ on the conversation and its instructions.
       countryCapitals.put("canada", "Ottawa");
       countryCapitals.put("france", "Paris");
       countryCapitals.put("japan", "Tokyo");
-    
+
       String result =
               countryCapitals.getOrDefault(
                       country.toLowerCase(), "Sorry, I couldn't find the capital for " + country + ".");
       return Map.of("result", result); // Tools must return a Map
     }
-    
+
     // Add the tool to the agent
     FunctionTool capitalTool = FunctionTool.create(experiment.getClass(), "getCapitalCity");
     LlmAgent capitalAgent =
@@ -269,10 +269,10 @@ For scenarios requiring structured data exchange with an `LLM Agent`, the ADK pr
 
     ```python
     from pydantic import BaseModel, Field
-    
+
     class CapitalOutput(BaseModel):
         capital: str = Field(description="The capital of the country.")
-    
+
     structured_capital_agent = LlmAgent(
         # ... name, model, description
         instruction="""You are a Capital Information Agent. Given a country, respond ONLY with a JSON object containing the capital. Format: {"capital": "capital_name"}""",
@@ -299,7 +299,7 @@ For scenarios requiring structured data exchange with an `LLM Agent`, the ADK pr
                         .description("The capital city of the country.")
                         .build()))
             .build();
-    
+
     LlmAgent structuredCapitalAgent =
         LlmAgent.builder()
             // ... name, model, description
@@ -332,14 +332,53 @@ Control whether the agent receives the prior conversation history.
 
     ```java
     import com.google.adk.agents.LlmAgent.IncludeContents;
-    
+
     LlmAgent statelessAgent =
         LlmAgent.builder()
             // ... other params
             .includeContents(IncludeContents.NONE)
             .build();
     ```
 
+### Context Compaction
+
+![python_only](https://img.shields.io/badge/Currently_supported_in-Python-blue){ title="This feature is currently available for Python. Java support is planned/ coming soon."}
+
+For long-running conversations, the history of events can grow very large, leading to increased latency and cost. The ADK provides a context compaction feature to mitigate this. It summarizes the conversation history at regular intervals, reducing the context size sent to the LLM.
+
+You can enable this feature by providing an `EventsCompactionConfig` when creating your `App`.
+
+* **`events_compaction_config` (Optional):** An instance of `EventsCompactionConfig` to configure context compaction.
+    * **`compaction_interval`:** The number of new user-initiated invocations that will trigger a compaction.
+    * **`summarizer`:** An instance of a `BaseEventsSummarizer`. The `LlmEventSummarizer` is a good default choice.
+    * **`overlap_size`:** The number of invocations to overlap between compaction windows to maintain context.
+
+=== "Python"
+
+    ```python
+    from google.adk.apps import App, EventsCompactionConfig
+    from google.adk.apps.llm_event_summarizer import LlmEventSummarizer
+    from google.adk.agents import LlmAgent
+
+    # Define your agent
+    my_agent = LlmAgent(
+        # ... your agent configuration
+    )
+
+    # Configure context compaction
+    compaction_config = EventsCompactionConfig(
+        compaction_interval=2,
+        summarizer=LlmEventSummarizer(model="gemini-1.5-flash"),
+        overlap_size=1
+    )
+
+    # Create an App with the agent and compaction config
+    my_app = App(
+        agent=my_agent,
+        events_compaction_config=compaction_config
+    )
+    ```
+
 ### Planner
 
 ![python_only](https://img.shields.io/badge/Currently_supported_in-Python-blue){ title="This feature is currently available for Python. Java support is planned/ coming soon."}
@@ -532,13 +571,13 @@ call_agent("If it's raining in New York right now, what is the current temperatu
     Here's the complete basic `capital_agent`:
 
     === "Python"
-    
+
         ```python
         --8<-- "examples/python/snippets/agents/llm-agent/capital_agent.py"
         ```
-    
+
     === "Java"
-    
+
         ```java
         --8<-- "examples/java/snippets/src/main/java/agents/LlmAgentExample.java:full_code"
         ```

docs/sessions/index.md

@@ -70,6 +70,14 @@ local testing, ADK also offers cloud-based and database service options.
     archive* potentially spanning across conversations. Managed by a
     `MemoryService`.
 
+## Context Compaction
+
+As conversations with an agent become longer, the amount of history (Events) that needs to be processed for each new turn can become very large. This can lead to increased latency and cost. To manage this, the ADK provides a feature called **Context Compaction**.
+
+Context Compaction works by summarizing older parts of the conversation history into a more compact form. This summary replaces the original events in the conversation history that is sent to the LLM, reducing the size of the context while still preserving the essential information from the past. This is particularly useful for long-running conversations where the agent needs to maintain context over many turns.
+
+You can configure context compaction at the application level, specifying how often it should be triggered and how the summarization should be performed.
+
 ## What's Next?
 
 In the following sections, we'll dive deeper into each of these components: