Add documentation for new examples (25-28)

- Agent delegation example
- Custom visualizer example
- Observability with Laminar
- Ask agent functionality

Co-authored-by: openhands <openhands@all-hands.dev>

Author: openhands-agent

sdk/guides/agent-delegation.mdx

@@ -0,0 +1,51 @@
+---
+title: Agent Delegation
+description: Delegate tasks to sub-agents for parallel processing
+---
+
+<Note>
+This example is available on GitHub: [examples/01_standalone_sdk/25_agent_delegation.py](https://github.com/All-Hands-AI/agent-sdk/blob/main/examples/01_standalone_sdk/25_agent_delegation.py)
+</Note>
+
+Learn how to use agent delegation to have a main agent delegate tasks to specialized sub-agents for parallel processing. Each sub-agent runs independently and returns results to the main agent for consolidation.
+
+```python icon="python" expandable examples/01_standalone_sdk/25_agent_delegation.py
+```
+
+```bash Running the Example
+export LLM_API_KEY="your-api-key"
+# Optional overrides
+# export LLM_MODEL="anthropic/claude-sonnet-4-5-20250929"
+# export LLM_BASE_URL="https://your-llm-base-url"
+
+cd agent-sdk
+uv run python examples/01_standalone_sdk/25_agent_delegation.py
+```
+
+### How It Works
+
+- Register the `DelegateTool` and create a delegation visualizer:
+
+```python
+from openhands.tools.delegate import DelegateTool, DelegationVisualizer, register_agent
+
+register_tool("DelegateTool", DelegateTool)
+visualizer = DelegationVisualizer()
+```
+
+- Register sub-agents with specific capabilities:
+
+```python
+register_agent(
+    agent_id="analyst",
+    agent=analyst_agent,
+    workspace=cwd,
+)
+```
+
+- The main agent can delegate tasks using the `DelegateTool`, and sub-agents will execute them independently.
+
+## Next Steps
+
+- **[Custom Tools](/sdk/guides/custom-tools)** – Create specialized tools for your agents
+- **[Multi-Agent Systems](/sdk/guides/multi-agent)** – Build complex agent hierarchies

sdk/guides/ask-agent.mdx

@@ -0,0 +1,57 @@
+---
+title: Ask Agent
+description: Get quick responses from agents without interrupting execution
+---
+
+<Note>
+This example is available on GitHub: [examples/01_standalone_sdk/28_ask_agent_example.py](https://github.com/All-Hands-AI/agent-sdk/blob/main/examples/01_standalone_sdk/28_ask_agent_example.py)
+</Note>
+
+Learn how to use `ask_agent()` to get quick responses from the agent about the current conversation state without interrupting the main execution flow.
+
+```python icon="python" expandable examples/01_standalone_sdk/28_ask_agent_example.py
+```
+
+```bash Running the Example
+export LLM_API_KEY="your-api-key"
+# Optional overrides
+# export LLM_MODEL="anthropic/claude-sonnet-4-5-20250929"
+# export LLM_BASE_URL="https://your-llm-base-url"
+
+cd agent-sdk
+uv run python examples/01_standalone_sdk/28_ask_agent_example.py
+```
+
+### How It Works
+
+- Create a conversation and start a long-running task:
+
+```python
+conversation = Conversation(agent=agent, workspace=".")
+conversation.send_message("Your main task message")
+```
+
+- While the conversation is running, use `ask_agent()` to get quick responses:
+
+```python
+response = conversation.ask_agent("What is the current status?")
+print(f"Agent response: {response}")
+```
+
+- The `ask_agent()` method:
+  - Does not interrupt the main conversation flow
+  - Returns a quick response based on the current conversation state
+  - Can be called multiple times during execution
+  - Useful for monitoring progress or getting status updates
+
+### Use Cases
+
+- **Progress monitoring**: Check how far along a task is
+- **Status queries**: Ask about current state or next steps
+- **Debug information**: Get insights into what the agent is doing
+- **Interactive monitoring**: Build UIs that show real-time updates
+
+## Next Steps
+
+- **[Conversation Management](/sdk/guides/conversation)** – Learn more about conversation APIs
+- **[Async Operations](/sdk/guides/async)** – Handle asynchronous agent interactions

sdk/guides/custom-visualizer.mdx

@@ -0,0 +1,57 @@
+---
+title: Custom Visualizer
+description: Create custom visualizers for conversation monitoring
+---
+
+<Note>
+This example is available on GitHub: [examples/01_standalone_sdk/26_custom_visualizer.py](https://github.com/All-Hands-AI/agent-sdk/blob/main/examples/01_standalone_sdk/26_custom_visualizer.py)
+</Note>
+
+Learn how to create custom visualizers by subclassing `ConversationVisualizerBase` to monitor and display conversation events in your preferred format.
+
+```python icon="python" expandable examples/01_standalone_sdk/26_custom_visualizer.py
+```
+
+```bash Running the Example
+export LLM_API_KEY="your-api-key"
+# Optional overrides
+# export LLM_MODEL="anthropic/claude-sonnet-4-5-20250929"
+# export LLM_BASE_URL="https://your-llm-base-url"
+
+cd agent-sdk
+uv run python examples/01_standalone_sdk/26_custom_visualizer.py
+```
+
+### How It Works
+
+- Create a custom visualizer by subclassing `ConversationVisualizerBase`:
+
+```python
+from openhands.sdk.conversation.visualizer import ConversationVisualizerBase
+from openhands.sdk.event import Event
+
+class MinimalVisualizer(ConversationVisualizerBase):
+    """A minimal visualizer that prints raw events."""
+    
+    def on_event(self, event: Event) -> None:
+        """Handle events for visualization."""
+        print(f"Event: {event}")
+```
+
+- Pass the visualizer instance to the conversation:
+
+```python
+visualizer = MinimalVisualizer()
+conversation = Conversation(
+    agent=agent,
+    workspace=".",
+    visualizer=visualizer,
+)
+```
+
+- The visualizer's `on_event` method will be called for each event in the conversation.
+
+## Next Steps
+
+- **[Event System](/sdk/guides/events)** – Learn about conversation events
+- **[Callbacks](/sdk/guides/callbacks)** – Use event callbacks for monitoring

sdk/guides/observability-laminar.mdx

@@ -0,0 +1,53 @@
+---
+title: Observability with Laminar
+description: Enable OpenTelemetry tracing with Laminar for observability
+---
+
+<Note>
+This example is available on GitHub: [examples/01_standalone_sdk/27_observability_laminar.py](https://github.com/All-Hands-AI/agent-sdk/blob/main/examples/01_standalone_sdk/27_observability_laminar.py)
+</Note>
+
+Learn how to enable OpenTelemetry tracing with Laminar to monitor and debug your agent conversations.
+
+```python icon="python" expandable examples/01_standalone_sdk/27_observability_laminar.py
+```
+
+```bash Running the Example
+export LLM_API_KEY="your-api-key"
+export LMNR_PROJECT_API_KEY="your-laminar-api-key"
+# Optional overrides
+# export LLM_MODEL="openhands/claude-sonnet-4-5-20250929"
+# export LLM_BASE_URL="https://your-llm-base-url"
+
+cd agent-sdk
+uv run python examples/01_standalone_sdk/27_observability_laminar.py
+```
+
+### How It Works
+
+- Set the `LMNR_PROJECT_API_KEY` environment variable before running:
+
+```bash
+export LMNR_PROJECT_API_KEY="your-laminar-api-key"
+```
+
+- The SDK automatically detects the Laminar configuration and enables OpenTelemetry tracing.
+
+- Create and run your conversation as usual:
+
+```python
+conversation = Conversation(agent=agent, workspace=".")
+conversation.send_message("List the files in the current directory.")
+conversation.run()
+```
+
+- Check your Laminar dashboard to see traces (the session ID matches the conversation UUID).
+
+### Using Other OTLP Backends
+
+For non-Laminar OpenTelemetry backends, set `OTEL_*` environment variables instead of `LMNR_PROJECT_API_KEY`.
+
+## Next Steps
+
+- **[Monitoring](/sdk/guides/monitoring)** – Learn about other monitoring options
+- **[Logging](/sdk/guides/logging)** – Configure logging for debugging