From a1e025e599c95bcfcb34d8640c1cde9c8bacb062 Mon Sep 17 00:00:00 2001
From: Arpandeep Khatua <arpandeepk@gmail.com>
Date: Tue, 12 Nov 2024 08:27:15 -0800
Subject: [PATCH 01/15] Adding OpenHands node

---
 .../group_discussion_agents.py                |   2 +-
 .../openhands_node/openhands_node.toml        |  11 +
 sotopia/experimental/__init__.py              |   2 +-
 .../{agents.py => agents/base_agent.py}       |  68 ++++-
 sotopia/experimental/nodes/openhands_node.py  | 245 ++++++++++++++++++
 5 files changed, 320 insertions(+), 8 deletions(-)
 create mode 100644 examples/experimental/openhands_node/openhands_node.toml
 rename sotopia/experimental/{agents.py => agents/base_agent.py} (53%)
 create mode 100644 sotopia/experimental/nodes/openhands_node.py

diff --git a/examples/experimental/group_discussion_agents/group_discussion_agents.py b/examples/experimental/group_discussion_agents/group_discussion_agents.py
index e4b3c0c2c..8ef55e5c5 100644
--- a/examples/experimental/group_discussion_agents/group_discussion_agents.py
+++ b/examples/experimental/group_discussion_agents/group_discussion_agents.py
@@ -2,7 +2,7 @@
 from aact import Message, NodeFactory
 from aact.messages import Text, Tick, DataModel, DataModelFactory
 from sotopia.agents.llm_agent import ainput
-from sotopia.experimental.agents import BaseAgent
+from sotopia.experimental.agents.base_agent import BaseAgent
 
 from sotopia.generation_utils import agenerate
 from sotopia.generation_utils.generate import StrOutputParser
diff --git a/examples/experimental/openhands_node/openhands_node.toml b/examples/experimental/openhands_node/openhands_node.toml
new file mode 100644
index 000000000..38ec6ab2e
--- /dev/null
+++ b/examples/experimental/openhands_node/openhands_node.toml
@@ -0,0 +1,11 @@
+redis_url = "redis://localhost:6379/0"
+extra_modules = ["sotopia.experimental.nodes.openhands_node"]
+
+
+[[nodes]]
+node_name = "runtime"
+node_class = "openhands"
+
+[nodes.node_args]
+output_channels = ['Runtime:Agent']
+input_channels = ['Agent:Runtime']
\ No newline at end of file
diff --git a/sotopia/experimental/__init__.py b/sotopia/experimental/__init__.py
index 21175f409..2f2166320 100644
--- a/sotopia/experimental/__init__.py
+++ b/sotopia/experimental/__init__.py
@@ -1,3 +1,3 @@
-from .agents import BaseAgent
+from .agents.base_agent import BaseAgent
 
 __all__ = ["BaseAgent"]
diff --git a/sotopia/experimental/agents.py b/sotopia/experimental/agents/base_agent.py
similarity index 53%
rename from sotopia/experimental/agents.py
rename to sotopia/experimental/agents/base_agent.py
index 209fc22d8..770a544c7 100644
--- a/sotopia/experimental/agents.py
+++ b/sotopia/experimental/agents/base_agent.py
@@ -1,19 +1,75 @@
 import asyncio
 import sys
 
+from enum import Enum
+from pydantic import Field
+
 if sys.version_info < (3, 11):
     from typing_extensions import Self
 else:
     from typing import Self
 
-from typing import Any, AsyncIterator, TypeVar
-from aact import Node
-from aact.messages import DataModel, Message
+from typing import Any, AsyncIterator, TypeVar, Optional
+
+from aact import Message, Node
+from aact.messages import DataModel
+from aact.messages.registry import DataModelFactory
 
 T_agent_observation = TypeVar("T_agent_observation", bound=DataModel)
 T_agent_action = TypeVar("T_agent_action", bound=DataModel)
 
 
+class ActionType(Enum):
+    NONE = "none"
+    SPEAK = "speak"
+    NON_VERBAL = "non-verbal"
+    LEAVE = "leave"
+    THOUGHT = "thought"
+    BROWSE = "browse"
+    BROWSE_ACTION = "browse_action"
+    READ = "read"
+    WRITE = "write"
+    RUN = "run"
+
+    def __str__(self) -> str:
+        return self.value
+
+    def __eq__(self, other: object) -> bool:
+        if isinstance(other, ActionType):
+            return self.value == other.value
+        elif isinstance(other, str):
+            return self.value == other
+        else:
+            return NotImplemented
+
+
+@DataModelFactory.register("agent_action")
+class AgentAction(DataModel):
+    agent_name: str = Field(description="the name of the agent")
+    action_type: ActionType = Field(
+        description="whether to speak at this turn or choose to not do anything"
+    )
+    argument: str = Field(
+        description="the utterance if choose to speak, the expression or gesture if choose non-verbal communication, or the physical action if choose action"
+    )
+    path: Optional[str] = Field(description="path of file")
+
+    def to_natural_language(self) -> str:
+        action_descriptions = {
+            ActionType.NONE: "did nothing",
+            ActionType.SPEAK: f'said: "{self.argument}"',
+            ActionType.THOUGHT: f'thought: "{self.argument}"',
+            ActionType.BROWSE: f'browsed: "{self.argument}"',
+            ActionType.RUN: f'ran: "{self.argument}"',
+            ActionType.READ: f'read: "{self.argument}"',
+            ActionType.WRITE: f'wrote: "{self.argument}"',
+            ActionType.NON_VERBAL: f"[{self.action_type.value}] {self.argument}",
+            ActionType.LEAVE: "left the conversation",
+        }
+
+        return action_descriptions.get(self.action_type, "performed an unknown action")
+
+
 class BaseAgent(Node[T_agent_observation, T_agent_action]):
     def __init__(
         self,
@@ -51,13 +107,13 @@ async def event_handler(
             await self.observation_queue.put(message.data)
         else:
             raise ValueError(f"Invalid channel: {channel}")
-            yield "", self.output_type()  # unreachable code
+            yield "", self.output_type() 
 
     async def send(self, action: T_agent_action) -> None:
         for output_channel, output_channel_type in self.output_channel_types.items():
             await self.r.publish(
                 output_channel,
-                Message[output_channel_type](data=action).model_dump_json(),  # type: ignore[valid-type]
+                Message[output_channel_type](data=action).model_dump_json(),
             )
 
     async def _task_scheduler(self) -> None:
@@ -66,4 +122,4 @@ async def _task_scheduler(self) -> None:
             action_or_none = await self.aact(observation)
             if action_or_none is not None:
                 await self.send(action_or_none)
-            self.observation_queue.task_done()
+            self.observation_queue.task_done()
\ No newline at end of file
diff --git a/sotopia/experimental/nodes/openhands_node.py b/sotopia/experimental/nodes/openhands_node.py
new file mode 100644
index 000000000..eea27032b
--- /dev/null
+++ b/sotopia/experimental/nodes/openhands_node.py
@@ -0,0 +1,245 @@
+import asyncio
+import logging
+import time
+import os
+import sys
+from typing import Any, AsyncIterator, Optional
+
+from rich.logging import RichHandler
+
+from aact import Message, NodeFactory, Node
+from aact.messages import Text, DataModel
+from aact.messages.commons import DataEntry
+
+from sotopia.experimental.agents.base_agent import AgentAction, ActionType
+
+from openhands.core.config import AgentConfig, AppConfig, SandboxConfig
+from openhands.core.logger import openhands_logger as logger
+from openhands.core.main import create_runtime
+from openhands.events.action import (
+    BrowseURLAction,
+    CmdRunAction,
+    FileWriteAction,
+    FileReadAction,
+    BrowseInteractiveAction,
+)
+from openhands.runtime.base import Runtime
+from openhands.utils.async_utils import call_async_from_sync
+
+if sys.version_info >= (3, 11):
+    from typing import Self
+else:
+    from typing_extensions import Self
+
+BASE_CONTAINER_IMAGE = "docker.all-hands.dev/all-hands-ai/runtime:0.11-nikolaik"
+
+# Configuration for logging
+FORMAT = "%(asctime)s - %(levelname)s - %(name)s - %(message)s"
+logging.basicConfig(
+    level=logging.WARNING,
+    format=FORMAT,
+    datefmt="[%X]",
+    handlers=[RichHandler()],
+)
+
+
+@NodeFactory.register("openhands")
+class OpenHands(Node[DataModel, Text]):
+    def __init__(
+        self,
+        input_channels: list[str],
+        output_channels: list[str],
+        redis_url: str,
+    ):
+        super().__init__(
+            input_channel_types=[
+                (input_channel, AgentAction) for input_channel in input_channels
+            ],
+            output_channel_types=[
+                (output_channel, Text) for output_channel in output_channels
+            ],
+            redis_url=redis_url,
+        )
+        self.queue: asyncio.Queue[DataEntry[DataModel]] = asyncio.Queue()
+        self.task: asyncio.Task[None] | None = None
+        self.runtime: Optional[Runtime] = None
+
+    async def init_runtime(self) -> None:
+        """
+        Initializes the runtime environment with the specified configuration.
+        """
+        start_time = time.time()
+        modal_api_token_id = os.environ.get("MODAL_API_TOKEN_ID", "")
+        modal_api_token_secret = os.environ.get("MODAL_API_TOKEN_SECRET", "")
+        allhands_api_key = os.environ.get("ALLHANDS_API_KEY", None)
+        sandbox_remote_runtime_api_url = os.environ.get(
+            "SANDBOX_REMOTE_RUNTIME_API_URL", ""
+        )
+
+        if not modal_api_token_id or not modal_api_token_secret:
+            logger.warning("Modal API tokens are not set. Check environment variables.")
+
+        config = AppConfig(
+            default_agent="CodeActAgent",
+            run_as_openhands=False,
+            max_iterations=3,
+            runtime="modal",
+            modal_api_token_id=modal_api_token_id,
+            modal_api_token_secret=modal_api_token_secret,
+            sandbox=SandboxConfig(
+                base_container_image=BASE_CONTAINER_IMAGE,
+                enable_auto_lint=True,
+                use_host_network=False,
+                timeout=50,
+                platform="linux/amd64",
+                api_key=allhands_api_key,
+                remote_runtime_api_url=sandbox_remote_runtime_api_url,
+                keep_remote_runtime_alive=False,
+            ),
+            workspace_base=None,
+            workspace_mount_path=None,
+        )
+
+        agent_config = AgentConfig(
+            codeact_enable_jupyter=True,
+            codeact_enable_browsing=True,
+            codeact_enable_llm_editor=True,
+        )
+        config.set_agent_config(agent_config)
+
+        self.runtime = create_runtime(config)
+        if self.runtime:
+            call_async_from_sync(self.runtime.connect)
+            logger.info("-" * 20)
+            logger.info("RUNTIME CONNECTED")
+            logger.info("-" * 20)
+        else:
+            logger.error("Failed to initialize runtime.")
+        end_time = time.time()  # End timing
+        elapsed_time = end_time - start_time
+        logger.info(f"Runtime initialization took {elapsed_time:.2f} seconds.")
+
+    async def __aenter__(self) -> Self:
+        self.runtime_init_task = asyncio.create_task(self.init_runtime())
+        self.task = asyncio.create_task(self.run_action())
+        return await super().__aenter__()
+
+    async def __aexit__(self, exc_type: Any, exc_value: Any, traceback: Any) -> None:
+        if self.runtime:
+            self.runtime.close()
+        return await super().__aexit__(exc_type, exc_value, traceback)
+
+    async def aact(self, action: AgentAction) -> Optional[Text]:
+        """
+        Executes an action based on the observation and returns the result as Text.
+
+        Args:
+            observation (AgentAction): The action to be executed.
+
+        Returns:
+            Optional[Text]: The result of the action, or None if the runtime is not available.
+        """
+        if not self.runtime:
+            logger.warning("Runtime is not initialized.")
+            return None
+
+        try:
+            action_obj = self._create_action(action)
+            logger.info(f"Executing action: {action}", extra={"msg_type": "ACTION"})
+            obs = self.runtime.run_action(action_obj)
+            logger.info(
+                f"Received observation: {str(obs).splitlines()[:2]}",
+                extra={"msg_type": "OBSERVATION"},
+            )
+            return Text(text=str(obs))
+        except Exception as e:
+            logger.error(f"Error executing action: {e}")
+            return None
+
+    def _create_action(self, observation: AgentAction) -> Any:
+        """
+        Creates an action based on the observation's action type.
+
+        Args:
+            observation (AgentAction): The observation containing the action type and arguments.
+
+        Returns:
+            Any: The created action.
+        """
+        action_type = observation.action_type
+        argument = observation.argument
+        path = observation.path
+
+        if action_type == ActionType.BROWSE:
+            return BrowseURLAction(url=argument)
+        elif action_type == ActionType.BROWSE_ACTION:
+            return BrowseInteractiveAction(browser_actions=argument)
+        elif action_type == ActionType.RUN:
+            return CmdRunAction(command=argument)
+        elif action_type == ActionType.WRITE:
+            if path is None:
+                raise ValueError("Path cannot be None for WRITE action")
+            return FileWriteAction(path=path, content=argument)
+        elif action_type == ActionType.READ:
+            if path is None:
+                raise ValueError("Path cannot be None for READ action")
+            return FileReadAction(path=path)
+        else:
+            raise ValueError(f"Unsupported action type: {action_type}")
+
+    async def send(self, action: Text) -> None:
+        """
+        Sends the action to all output channels.
+
+        Args:
+            action (Text): The action to be sent.
+        """
+        try:
+            for output_channel, _ in self.output_channel_types.items():
+                message = Message[Text](data=action).model_dump_json()
+                await self.r.publish(output_channel, message)
+        except Exception as e:
+            logger.error(f"Error sending action: {e}")
+
+    async def run_action(self) -> None:
+        """
+        Continuously processes actions from the queue.
+        """
+        while self.task:
+            try:
+                data_entry = await self.queue.get()
+                if isinstance(data_entry.data, AgentAction):
+                    obs = await self.aact(data_entry.data)
+                    if obs is not None:
+                        await self.send(obs)
+                else:
+                    logger.error("Data is not of type AgentAction")
+                self.queue.task_done()
+            except Exception as e:
+                logger.error(f"Error processing action: {e}")
+
+    async def event_handler(
+        self, input_channel: str, input_message: Message[DataModel]
+    ) -> AsyncIterator[tuple[str, Message[Text]]]:
+        """
+        Handles incoming events and adds them to the processing queue.
+
+        Args:
+            input_channel (str): The channel from which the message was received.
+            input_message (Message[DataModel]): The incoming message.
+
+        Yields:
+            Tuple[str, Message[Zero]]: A tuple containing the channel and a zero message if the channel is not recognized.
+        """
+        try:
+            if input_channel in self.input_channel_types:
+                # Create a DataEntry instance with the correct type
+                data_entry: DataEntry[DataModel] = DataEntry(
+                    channel=input_channel, data=input_message.data
+                )
+                await self.queue.put(data_entry)
+            else:
+                logger.warning(f"Unrecognized input channel: {input_channel}")
+                yield input_channel, Message[Text](data=Text(text=""))
+        except Exception as e:
+            logger.error(f"Error handling event: {e}")
\ No newline at end of file

From ae72e44e6d784f39b8a2b20c97162331372b3a52 Mon Sep 17 00:00:00 2001
From: "autofix-ci[bot]" <114827586+autofix-ci[bot]@users.noreply.github.com>
Date: Tue, 12 Nov 2024 16:29:31 +0000
Subject: [PATCH 02/15] [autofix.ci] apply automated fixes

---
 examples/experimental/openhands_node/openhands_node.toml | 2 +-
 sotopia/experimental/agents/base_agent.py                | 4 ++--
 sotopia/experimental/nodes/openhands_node.py             | 2 +-
 3 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/examples/experimental/openhands_node/openhands_node.toml b/examples/experimental/openhands_node/openhands_node.toml
index 38ec6ab2e..a3987fea5 100644
--- a/examples/experimental/openhands_node/openhands_node.toml
+++ b/examples/experimental/openhands_node/openhands_node.toml
@@ -8,4 +8,4 @@ node_class = "openhands"
 
 [nodes.node_args]
 output_channels = ['Runtime:Agent']
-input_channels = ['Agent:Runtime']
\ No newline at end of file
+input_channels = ['Agent:Runtime']
diff --git a/sotopia/experimental/agents/base_agent.py b/sotopia/experimental/agents/base_agent.py
index 770a544c7..dca34e5e0 100644
--- a/sotopia/experimental/agents/base_agent.py
+++ b/sotopia/experimental/agents/base_agent.py
@@ -107,7 +107,7 @@ async def event_handler(
             await self.observation_queue.put(message.data)
         else:
             raise ValueError(f"Invalid channel: {channel}")
-            yield "", self.output_type() 
+            yield "", self.output_type()
 
     async def send(self, action: T_agent_action) -> None:
         for output_channel, output_channel_type in self.output_channel_types.items():
@@ -122,4 +122,4 @@ async def _task_scheduler(self) -> None:
             action_or_none = await self.aact(observation)
             if action_or_none is not None:
                 await self.send(action_or_none)
-            self.observation_queue.task_done()
\ No newline at end of file
+            self.observation_queue.task_done()
diff --git a/sotopia/experimental/nodes/openhands_node.py b/sotopia/experimental/nodes/openhands_node.py
index eea27032b..59671d7b0 100644
--- a/sotopia/experimental/nodes/openhands_node.py
+++ b/sotopia/experimental/nodes/openhands_node.py
@@ -242,4 +242,4 @@ async def event_handler(
                 logger.warning(f"Unrecognized input channel: {input_channel}")
                 yield input_channel, Message[Text](data=Text(text=""))
         except Exception as e:
-            logger.error(f"Error handling event: {e}")
\ No newline at end of file
+            logger.error(f"Error handling event: {e}")

From dfeaa82708df100731d6ace324bab9cf2887c3ef Mon Sep 17 00:00:00 2001
From: Arpandeep Khatua <arpandeepk@gmail.com>
Date: Tue, 12 Nov 2024 09:22:34 -0800
Subject: [PATCH 03/15] Added LLM Agent Node

---
 .../experimental/llm_agent/llm_agent.toml     |  45 ++
 sotopia/experimental/agents/llm_agent.py      | 390 ++++++++++++++++++
 2 files changed, 435 insertions(+)
 create mode 100644 examples/experimental/llm_agent/llm_agent.toml
 create mode 100644 sotopia/experimental/agents/llm_agent.py

diff --git a/examples/experimental/llm_agent/llm_agent.toml b/examples/experimental/llm_agent/llm_agent.toml
new file mode 100644
index 000000000..a204bccd0
--- /dev/null
+++ b/examples/experimental/llm_agent/llm_agent.toml
@@ -0,0 +1,45 @@
+redis_url = "redis://localhost:6379/0"
+extra_modules = ["sotopia.experimental.agents.llm_agent"]
+
+
+[[nodes]]
+node_name = "Jack"
+node_class = "llm_agent"
+
+[nodes.node_args]
+query_interval = 5
+output_channel = "Jack:Jane"
+input_text_channels = ["Jane:Jack", "Runtime:Agent"]
+input_env_channels = []
+input_tick_channel = "tick/secs/1"
+goal = "Your goal is to effectively test Jane's technical ability and finally decide if she has passed the interview. Make sure to also evaluate her communication skills, problem-solving approach, and enthusiasm."
+model_name = "gpt-4o-mini"
+agent_name = "Jack"
+
+[[nodes]]
+node_name = "Jane"
+node_class = "llm_agent"
+
+[nodes.node_args]
+query_interval = 7
+output_channel = "Jane:Jack"
+input_text_channels = ["Jack:Jane", "Runtime:Agent"]
+input_env_channels = []
+input_tick_channel = "tick/secs/1"
+goal = "Your goal is to do well in the interview by demonstrating your technical skills, clear communication, and enthusiasm for the position. Stay calm, ask clarifying questions when needed, and confidently explain your thought process."
+model_name = "gpt-4o-mini"
+agent_name = "Jane"
+
+[[nodes]]
+node_name = "tick"
+node_class = "tick"
+
+
+[[nodes]]
+node_name = "print"
+node_class = "print"
+
+[nodes.node_args.print_channel_types]
+"tick/secs/1" = "tick"
+"Jane:Jack" = "agent_action"
+"Jack:Jane" = "agent_action"
\ No newline at end of file
diff --git a/sotopia/experimental/agents/llm_agent.py b/sotopia/experimental/agents/llm_agent.py
new file mode 100644
index 000000000..8cb1fc428
--- /dev/null
+++ b/sotopia/experimental/agents/llm_agent.py
@@ -0,0 +1,390 @@
+import logging
+import sys
+
+from rich.logging import RichHandler
+
+from aact import Message, NodeFactory
+from aact.messages import Text, Tick
+from sotopia.experimental.agents.base_agent import BaseAgent, AgentAction, ActionType
+
+from sotopia.generation_utils import agenerate
+from sotopia.generation_utils.generate import StrOutputParser
+
+import json
+import traceback
+
+# Check Python version
+if sys.version_info >= (3, 11):
+    pass
+else:
+    pass
+
+# Configure logging
+FORMAT = "%(asctime)s - %(levelname)s - %(name)s - %(message)s"
+logging.basicConfig(
+    level=logging.WARNING,
+    format=FORMAT,
+    datefmt="[%X]",
+    handlers=[RichHandler()],
+)
+
+
+@NodeFactory.register("llm_agent")
+class LLMAgent(BaseAgent[AgentAction | Tick | Text, AgentAction]):
+    def __init__(
+        self,
+        input_text_channels: list[str],
+        input_tick_channel: str,
+        input_env_channels: list[str],
+        output_channel: str,
+        query_interval: int,
+        agent_name: str,
+        goal: str,
+        model_name: str,
+        redis_url: str,
+    ):
+        super().__init__(
+            [
+                (input_text_channel, AgentAction)
+                for input_text_channel in input_text_channels
+            ]
+            + [
+                (input_tick_channel, Tick),
+            ]
+            + [(input_env_channel, Text) for input_env_channel in input_env_channels],
+            [(output_channel, AgentAction)],
+            redis_url,
+        )
+        self.output_channel = output_channel
+        self.query_interval = query_interval
+        self.count_ticks = 0
+        self.message_history: list[tuple[str, str, str]] = []
+        self.name = agent_name
+        self.model_name = model_name
+        self.goal = goal
+
+    async def send(self, message: AgentAction) -> None:
+        if message.action_type == "speak":
+            await self.r.publish(
+                self.output_channel,
+                Message[AgentAction](data=message).model_dump_json(),
+            )
+
+        elif message.action_type in ("browse", "browse_action", "write", "read", "run"):
+            await self.r.publish(
+                "Agent:Runtime",
+                Message[AgentAction](data=message).model_dump_json(),
+            )
+
+    def _format_message_history(
+        self, message_history: list[tuple[str, str, str]]
+    ) -> str:
+        ## TODO: akhatua Fix the mapping of action to be gramatically correct
+        return "\n".join(
+            (f"{speaker} {action} {message}")
+            for speaker, action, message in message_history
+        )
+
+    def get_action_template(self, selected_actions: list[ActionType]) -> str:
+        """
+        Returns the action template string with selected actions.
+
+        Args:
+            selected_actions (list[ActionType]): List of ActionType enum members to include in the template.
+
+        Returns:
+            str: The action template with the selected actions.
+        """
+        base_template = """ You are talking to another agent.
+        You are {agent_name}.\n
+        {message_history}\nand you plan to {goal}.
+        ## Action
+        What is your next thought or action? Your response must be in JSON format.
+
+        It must be an object, and it must contain two fields:
+        * `action`, which is one of the actions below
+        * `args`, which is a map of key-value pairs, specifying the arguments for that action
+        """
+
+        action_descriptions = {
+            str(
+                ActionType.SPEAK
+            ): """`speak` - you can talk to the other agents to share information or ask them something. Arguments:
+                * `content` - the message to send to the other agents (should be short)""",
+            str(
+                ActionType.THOUGHT
+            ): """`thought` - only use this rarely to make a plan, set a goal, record your thoughts. Arguments:
+                * `content` - the message you send yourself to organize your thoughts (should be short). You cannot think more than 2 turns.""",
+            str(
+                ActionType.NONE
+            ): """`none` - you can choose not to take an action if you are waiting for some data""",
+            str(
+                ActionType.NON_VERBAL
+            ): """`non-verbal` - you can choose to do a non verbal action
+                * `content` - the non veral action you want to send to other agents. eg: smile, shrug, thumbs up""",
+            str(ActionType.BROWSE): """`browse` - opens a web page. Arguments:
+                * `url` - the URL to open, when you browse the web you must use `none` action until you get some information back. When you get the information back you must summarize the article and explain the article to the other agents.""",
+            str(
+                ActionType.BROWSE_ACTION
+            ): """`browse_action` - actions you can take on a web browser
+                * `command` - the command to run. You have 15 available commands. These commands must be a single string value of command
+                    Options for `command`:
+                        `command` = goto(url: str)
+                            Description: Navigate to a url.
+                            Examples:
+                                goto('http://www.example.com')
+
+                        `command` = go_back()
+                            Description: Navigate to the previous page in history.
+                            Examples:
+                                go_back()
+
+                        `command` = go_forward()
+                            Description: Navigate to the next page in history.
+                            Examples:
+                                go_forward()
+
+                        `command` = noop(wait_ms: float = 1000)
+                            Description: Do nothing, and optionally wait for the given time (in milliseconds).
+                            You can use this to get the current page content and/or wait for the page to load.
+                            Examples:
+                                noop()
+                                noop(500)
+
+                        `command` = scroll(delta_x: float, delta_y: float)
+                            Description: Scroll horizontally and vertically. Amounts in pixels, positive for right or down scrolling, negative for left or up scrolling. Dispatches a wheel event.
+                            Examples:
+                                scroll(0, 200)
+                                scroll(-50.2, -100.5)
+
+                        `command` = fill(bid, value)
+                            Description: Fill out a form field. It focuses the element and triggers an input event with the entered text. It works for <input>, <textarea> and [contenteditable] elements.
+                            Examples:
+                                fill('237', 'example value')
+                                fill('45', 'multi-line\nexample')
+                                fill('a12', 'example with "quotes"')
+
+                        `command` = select_option(bid: str, options: str | list[str])
+                            Description: Select one or multiple options in a <select> element. You can specify option value or label to select. Multiple options can be selected.
+                            Examples:
+                                select_option('a48', 'blue')
+                                select_option('c48', ['red', 'green', 'blue'])
+
+                        `command`= click(bid: str, button: Literal['left', 'middle', 'right'] = 'left', modifiers: list[typing.Literal['Alt', 'Control', 'ControlOrMeta', 'Meta', 'Shift']] = [])
+                            Description: Click an element.
+                            Examples:
+                                click('a51')
+                                click('b22', button='right')
+                                click('48', button='middle', modifiers=['Shift'])
+
+                        `command` = dblclick(bid: str, button: Literal['left', 'middle', 'right'] = 'left', modifiers: list[typing.Literal['Alt', 'Control', 'ControlOrMeta', 'Meta', 'Shift']] = [])
+                            Description: Double click an element.
+                            Examples:
+                                dblclick('12')
+                                dblclick('ca42', button='right')
+                                dblclick('178', button='middle', modifiers=['Shift'])
+
+                        `command` = hover(bid: str)
+                            Description: Hover over an element.
+                            Examples:
+                                hover('b8')
+
+                        `command` = press(bid: str, key_comb: str)
+                            Description: Focus the matching element and press a combination of keys. It accepts the logical key names that are emitted in the keyboardEvent.key property of the keyboard events: Backquote, Minus, Equal, Backslash, Backspace, Tab, Delete, Escape, ArrowDown, End, Enter, Home, Insert, PageDown, PageUp, ArrowRight, ArrowUp, F1 - F12, Digit0 - Digit9, KeyA - KeyZ, etc. You can alternatively specify a single character you'd like to produce such as "a" or "#". Following modification shortcuts are also supported: Shift, Control, Alt, Meta, ShiftLeft, ControlOrMeta. ControlOrMeta resolves to Control on Windows and Linux and to Meta on macOS.
+                            Examples:
+                                press('88', 'Backspace')
+                                press('a26', 'ControlOrMeta+a')
+                                press('a61', 'Meta+Shift+t')
+
+                        `command` = focus(bid: str)
+                            Description: Focus the matching element.
+                            Examples:
+                                focus('b455')
+
+                        `command` = clear(bid: str)
+                            Description: Clear the input field.
+                            Examples:
+                                clear('996')
+
+                        `command` = drag_and_drop(from_bid: str, to_bid: str)
+                            Description: Perform a drag & drop. Hover the element that will be dragged. Press left mouse button. Move mouse to the element that will receive the drop. Release left mouse button.
+                            Examples:
+                                drag_and_drop('56', '498')
+
+                        `command`=  upload_file(bid: str, file: str | list[str])
+                            Description: Click an element and wait for a "filechooser" event, then select one or multiple input files for upload. Relative file paths are resolved relative to the current working directory. An empty list clears the selected files.
+                            Examples:
+                                upload_file('572', '/home/user/my_receipt.pdf')
+                                upload_file('63', ['/home/bob/Documents/image.jpg', '/home/bob/Documents/file.zip'])""",
+            str(ActionType.READ): """`read` - reads the content of a file. Arguments:
+                * `path` - the path of the file to read""",
+            str(ActionType.WRITE): """`write` - writes the content to a file. Arguments:
+                * `path` - the path of the file to write
+                * `content` - the content to write to the file""",
+            str(
+                ActionType.RUN
+            ): """`run` - runs a command on the command line in a Linux shell. Arguments:
+                * `command` - the command to run""",
+            str(
+                ActionType.LEAVE
+            ): """`leave` - if your goals have been completed or abandoned, and you're absolutely certain that you've completed your task and have tested your work, use the leave action to stop working.""",
+        }
+
+        selected_action_descriptions = "\n\n".join(
+            f"[{i+1}] {action_descriptions[str(action)]}"
+            for i, action in enumerate(selected_actions)
+            if str(action) in action_descriptions
+        )
+
+        return (
+            base_template
+            + selected_action_descriptions
+            + """
+                You must prioritize actions that move you closer to your goal. Communicate briefly when necessary and focus on executing tasks effectively. Always consider the next actionable step to avoid unnecessary delays.
+                Again, you must reply with JSON, and only with JSON.
+            """
+        )
+
+    async def aact(self, message: AgentAction | Tick | Text) -> AgentAction:
+        match message:
+            case Text(text=text):
+                self.message_history.append((self.name, "observation data", text))
+                return AgentAction(
+                    agent_name=self.name, action_type="none", argument="", path=""
+                )
+            case Tick():
+                self.count_ticks += 1
+                if self.count_ticks % self.query_interval == 0:
+                    template = self.get_action_template(
+                        [action for action in ActionType]
+                    )
+
+                    agent_action = await agenerate(
+                        model_name=self.model_name,
+                        template=template,
+                        input_values={
+                            "message_history": self._format_message_history(
+                                self.message_history
+                            ),
+                            "goal": self.goal,
+                            "agent_name": self.name,
+                        },
+                        temperature=0.7,
+                        output_parser=StrOutputParser(),
+                    )
+  
+                    agent_action = (
+                        agent_action.replace("```", "")
+                        .replace("json", "")
+                        .strip('"')
+                        .strip()
+                    )
+
+                    try:
+                        data = json.loads(agent_action)
+                        action = data["action"]
+                        if action == "thought":
+                            content = data["args"]["content"]
+                            self.message_history.append((self.name, action, content))
+                            return AgentAction(
+                                agent_name=self.name,
+                                action_type="thought",
+                                argument=content,
+                                path="",
+                            )
+
+                        elif action == "speak":
+                            content = data["args"]["content"]
+                            self.message_history.append((self.name, action, content))
+                            return AgentAction(
+                                agent_name=self.name,
+                                action_type=action,
+                                argument=content,
+                                path="",
+                            )
+
+                        elif action == "non-verbal":
+                            content = data["args"]["content"]
+                            self.message_history.append((self.name, action, content))
+                            return AgentAction(
+                                agent_name=self.name,
+                                action_type=action,
+                                argument=content,
+                                path="",
+                            )
+
+                        elif action == "browse":
+                            url = data["args"]["url"]
+                            self.message_history.append((self.name, action, url))
+                            return AgentAction(
+                                agent_name=self.name,
+                                action_type=action,
+                                argument=url,
+                                path="",
+                            )
+
+                        elif action == "browse_action":
+                            command = data["args"]["command"]
+                            self.message_history.append((self.name, action, command))
+                            return AgentAction(
+                                agent_name=self.name,
+                                action_type=action,
+                                argument=command,
+                                path="",
+                            )
+
+                        elif action == "run":
+                            command = data["args"]["command"]
+                            self.message_history.append((self.name, action, command))
+                            return AgentAction(
+                                agent_name=self.name,
+                                action_type=action,
+                                argument=command,
+                                path="",
+                            )
+
+                        elif action == "write":
+                            path = data["args"]["path"]
+                            content = data["args"]["content"]
+                            self.message_history.append((self.name, action, content))
+                            return AgentAction(
+                                agent_name=self.name,
+                                action_type=action,
+                                argument=content,
+                                path=path,
+                            )
+
+                        elif action == "read":
+                            path = data["args"]["path"]
+                            self.message_history.append((self.name, action, path))
+                            return AgentAction(
+                                agent_name=self.name,
+                                action_type=action,
+                                argument="Nan",
+                                path=path,
+                            )
+
+                        elif action == "none":
+                            return AgentAction(
+                                agent_name=self.name,
+                                action_type="none",
+                                argument="",
+                                path="",
+                            )
+                        else:
+                            print(f"Unknown action: {action}")
+                    except json.JSONDecodeError as e:
+                        print(f"Error decoding JSON: {e}")
+                else:
+                    return AgentAction(
+                        agent_name=self.name, action_type="none", argument="", path=""
+                    )
+            case AgentAction(
+                agent_name=agent_name, action_type=action_type, argument=text
+            ):
+                if action_type == "speak":
+                    self.message_history.append((agent_name, str(action_type), text))
+                return AgentAction(
+                    agent_name=self.name, action_type="none", argument="", path=""
+                )
+        raise ValueError(f"Unexpected message type: {type(message)}")
\ No newline at end of file

From 879da885b09663cbb69b68b66e8d3b737d2ac2ae Mon Sep 17 00:00:00 2001
From: "autofix-ci[bot]" <114827586+autofix-ci[bot]@users.noreply.github.com>
Date: Tue, 12 Nov 2024 17:23:20 +0000
Subject: [PATCH 04/15] [autofix.ci] apply automated fixes

---
 examples/experimental/llm_agent/llm_agent.toml | 2 +-
 sotopia/experimental/agents/llm_agent.py       | 5 ++---
 2 files changed, 3 insertions(+), 4 deletions(-)

diff --git a/examples/experimental/llm_agent/llm_agent.toml b/examples/experimental/llm_agent/llm_agent.toml
index a204bccd0..63061a798 100644
--- a/examples/experimental/llm_agent/llm_agent.toml
+++ b/examples/experimental/llm_agent/llm_agent.toml
@@ -42,4 +42,4 @@ node_class = "print"
 [nodes.node_args.print_channel_types]
 "tick/secs/1" = "tick"
 "Jane:Jack" = "agent_action"
-"Jack:Jane" = "agent_action"
\ No newline at end of file
+"Jack:Jane" = "agent_action"
diff --git a/sotopia/experimental/agents/llm_agent.py b/sotopia/experimental/agents/llm_agent.py
index 8cb1fc428..789deb8a6 100644
--- a/sotopia/experimental/agents/llm_agent.py
+++ b/sotopia/experimental/agents/llm_agent.py
@@ -11,7 +11,6 @@
 from sotopia.generation_utils.generate import StrOutputParser
 
 import json
-import traceback
 
 # Check Python version
 if sys.version_info >= (3, 11):
@@ -272,7 +271,7 @@ async def aact(self, message: AgentAction | Tick | Text) -> AgentAction:
                         temperature=0.7,
                         output_parser=StrOutputParser(),
                     )
-  
+
                     agent_action = (
                         agent_action.replace("```", "")
                         .replace("json", "")
@@ -387,4 +386,4 @@ async def aact(self, message: AgentAction | Tick | Text) -> AgentAction:
                 return AgentAction(
                     agent_name=self.name, action_type="none", argument="", path=""
                 )
-        raise ValueError(f"Unexpected message type: {type(message)}")
\ No newline at end of file
+        raise ValueError(f"Unexpected message type: {type(message)}")

From d15509d056a06385f55bec0c70a76c97316bc97c Mon Sep 17 00:00:00 2001
From: Arpandeep Khatua <arpandeepk@gmail.com>
Date: Tue, 12 Nov 2024 17:06:54 -0800
Subject: [PATCH 05/15] removing openhands

---
 .../openhands_node/openhands_node.toml        |  11 -
 sotopia/experimental/agents/base_agent.py     |   2 +-
 sotopia/experimental/nodes/openhands_node.py  | 245 ------------------
 3 files changed, 1 insertion(+), 257 deletions(-)
 delete mode 100644 examples/experimental/openhands_node/openhands_node.toml
 delete mode 100644 sotopia/experimental/nodes/openhands_node.py

diff --git a/examples/experimental/openhands_node/openhands_node.toml b/examples/experimental/openhands_node/openhands_node.toml
deleted file mode 100644
index a3987fea5..000000000
--- a/examples/experimental/openhands_node/openhands_node.toml
+++ /dev/null
@@ -1,11 +0,0 @@
-redis_url = "redis://localhost:6379/0"
-extra_modules = ["sotopia.experimental.nodes.openhands_node"]
-
-
-[[nodes]]
-node_name = "runtime"
-node_class = "openhands"
-
-[nodes.node_args]
-output_channels = ['Runtime:Agent']
-input_channels = ['Agent:Runtime']
diff --git a/sotopia/experimental/agents/base_agent.py b/sotopia/experimental/agents/base_agent.py
index dca34e5e0..4202f740e 100644
--- a/sotopia/experimental/agents/base_agent.py
+++ b/sotopia/experimental/agents/base_agent.py
@@ -113,7 +113,7 @@ async def send(self, action: T_agent_action) -> None:
         for output_channel, output_channel_type in self.output_channel_types.items():
             await self.r.publish(
                 output_channel,
-                Message[output_channel_type](data=action).model_dump_json(),
+                Message[output_channel_type](data=action).model_dump_json(), # type:ignore[valid-type] 
             )
 
     async def _task_scheduler(self) -> None:
diff --git a/sotopia/experimental/nodes/openhands_node.py b/sotopia/experimental/nodes/openhands_node.py
deleted file mode 100644
index 59671d7b0..000000000
--- a/sotopia/experimental/nodes/openhands_node.py
+++ /dev/null
@@ -1,245 +0,0 @@
-import asyncio
-import logging
-import time
-import os
-import sys
-from typing import Any, AsyncIterator, Optional
-
-from rich.logging import RichHandler
-
-from aact import Message, NodeFactory, Node
-from aact.messages import Text, DataModel
-from aact.messages.commons import DataEntry
-
-from sotopia.experimental.agents.base_agent import AgentAction, ActionType
-
-from openhands.core.config import AgentConfig, AppConfig, SandboxConfig
-from openhands.core.logger import openhands_logger as logger
-from openhands.core.main import create_runtime
-from openhands.events.action import (
-    BrowseURLAction,
-    CmdRunAction,
-    FileWriteAction,
-    FileReadAction,
-    BrowseInteractiveAction,
-)
-from openhands.runtime.base import Runtime
-from openhands.utils.async_utils import call_async_from_sync
-
-if sys.version_info >= (3, 11):
-    from typing import Self
-else:
-    from typing_extensions import Self
-
-BASE_CONTAINER_IMAGE = "docker.all-hands.dev/all-hands-ai/runtime:0.11-nikolaik"
-
-# Configuration for logging
-FORMAT = "%(asctime)s - %(levelname)s - %(name)s - %(message)s"
-logging.basicConfig(
-    level=logging.WARNING,
-    format=FORMAT,
-    datefmt="[%X]",
-    handlers=[RichHandler()],
-)
-
-
-@NodeFactory.register("openhands")
-class OpenHands(Node[DataModel, Text]):
-    def __init__(
-        self,
-        input_channels: list[str],
-        output_channels: list[str],
-        redis_url: str,
-    ):
-        super().__init__(
-            input_channel_types=[
-                (input_channel, AgentAction) for input_channel in input_channels
-            ],
-            output_channel_types=[
-                (output_channel, Text) for output_channel in output_channels
-            ],
-            redis_url=redis_url,
-        )
-        self.queue: asyncio.Queue[DataEntry[DataModel]] = asyncio.Queue()
-        self.task: asyncio.Task[None] | None = None
-        self.runtime: Optional[Runtime] = None
-
-    async def init_runtime(self) -> None:
-        """
-        Initializes the runtime environment with the specified configuration.
-        """
-        start_time = time.time()
-        modal_api_token_id = os.environ.get("MODAL_API_TOKEN_ID", "")
-        modal_api_token_secret = os.environ.get("MODAL_API_TOKEN_SECRET", "")
-        allhands_api_key = os.environ.get("ALLHANDS_API_KEY", None)
-        sandbox_remote_runtime_api_url = os.environ.get(
-            "SANDBOX_REMOTE_RUNTIME_API_URL", ""
-        )
-
-        if not modal_api_token_id or not modal_api_token_secret:
-            logger.warning("Modal API tokens are not set. Check environment variables.")
-
-        config = AppConfig(
-            default_agent="CodeActAgent",
-            run_as_openhands=False,
-            max_iterations=3,
-            runtime="modal",
-            modal_api_token_id=modal_api_token_id,
-            modal_api_token_secret=modal_api_token_secret,
-            sandbox=SandboxConfig(
-                base_container_image=BASE_CONTAINER_IMAGE,
-                enable_auto_lint=True,
-                use_host_network=False,
-                timeout=50,
-                platform="linux/amd64",
-                api_key=allhands_api_key,
-                remote_runtime_api_url=sandbox_remote_runtime_api_url,
-                keep_remote_runtime_alive=False,
-            ),
-            workspace_base=None,
-            workspace_mount_path=None,
-        )
-
-        agent_config = AgentConfig(
-            codeact_enable_jupyter=True,
-            codeact_enable_browsing=True,
-            codeact_enable_llm_editor=True,
-        )
-        config.set_agent_config(agent_config)
-
-        self.runtime = create_runtime(config)
-        if self.runtime:
-            call_async_from_sync(self.runtime.connect)
-            logger.info("-" * 20)
-            logger.info("RUNTIME CONNECTED")
-            logger.info("-" * 20)
-        else:
-            logger.error("Failed to initialize runtime.")
-        end_time = time.time()  # End timing
-        elapsed_time = end_time - start_time
-        logger.info(f"Runtime initialization took {elapsed_time:.2f} seconds.")
-
-    async def __aenter__(self) -> Self:
-        self.runtime_init_task = asyncio.create_task(self.init_runtime())
-        self.task = asyncio.create_task(self.run_action())
-        return await super().__aenter__()
-
-    async def __aexit__(self, exc_type: Any, exc_value: Any, traceback: Any) -> None:
-        if self.runtime:
-            self.runtime.close()
-        return await super().__aexit__(exc_type, exc_value, traceback)
-
-    async def aact(self, action: AgentAction) -> Optional[Text]:
-        """
-        Executes an action based on the observation and returns the result as Text.
-
-        Args:
-            observation (AgentAction): The action to be executed.
-
-        Returns:
-            Optional[Text]: The result of the action, or None if the runtime is not available.
-        """
-        if not self.runtime:
-            logger.warning("Runtime is not initialized.")
-            return None
-
-        try:
-            action_obj = self._create_action(action)
-            logger.info(f"Executing action: {action}", extra={"msg_type": "ACTION"})
-            obs = self.runtime.run_action(action_obj)
-            logger.info(
-                f"Received observation: {str(obs).splitlines()[:2]}",
-                extra={"msg_type": "OBSERVATION"},
-            )
-            return Text(text=str(obs))
-        except Exception as e:
-            logger.error(f"Error executing action: {e}")
-            return None
-
-    def _create_action(self, observation: AgentAction) -> Any:
-        """
-        Creates an action based on the observation's action type.
-
-        Args:
-            observation (AgentAction): The observation containing the action type and arguments.
-
-        Returns:
-            Any: The created action.
-        """
-        action_type = observation.action_type
-        argument = observation.argument
-        path = observation.path
-
-        if action_type == ActionType.BROWSE:
-            return BrowseURLAction(url=argument)
-        elif action_type == ActionType.BROWSE_ACTION:
-            return BrowseInteractiveAction(browser_actions=argument)
-        elif action_type == ActionType.RUN:
-            return CmdRunAction(command=argument)
-        elif action_type == ActionType.WRITE:
-            if path is None:
-                raise ValueError("Path cannot be None for WRITE action")
-            return FileWriteAction(path=path, content=argument)
-        elif action_type == ActionType.READ:
-            if path is None:
-                raise ValueError("Path cannot be None for READ action")
-            return FileReadAction(path=path)
-        else:
-            raise ValueError(f"Unsupported action type: {action_type}")
-
-    async def send(self, action: Text) -> None:
-        """
-        Sends the action to all output channels.
-
-        Args:
-            action (Text): The action to be sent.
-        """
-        try:
-            for output_channel, _ in self.output_channel_types.items():
-                message = Message[Text](data=action).model_dump_json()
-                await self.r.publish(output_channel, message)
-        except Exception as e:
-            logger.error(f"Error sending action: {e}")
-
-    async def run_action(self) -> None:
-        """
-        Continuously processes actions from the queue.
-        """
-        while self.task:
-            try:
-                data_entry = await self.queue.get()
-                if isinstance(data_entry.data, AgentAction):
-                    obs = await self.aact(data_entry.data)
-                    if obs is not None:
-                        await self.send(obs)
-                else:
-                    logger.error("Data is not of type AgentAction")
-                self.queue.task_done()
-            except Exception as e:
-                logger.error(f"Error processing action: {e}")
-
-    async def event_handler(
-        self, input_channel: str, input_message: Message[DataModel]
-    ) -> AsyncIterator[tuple[str, Message[Text]]]:
-        """
-        Handles incoming events and adds them to the processing queue.
-
-        Args:
-            input_channel (str): The channel from which the message was received.
-            input_message (Message[DataModel]): The incoming message.
-
-        Yields:
-            Tuple[str, Message[Zero]]: A tuple containing the channel and a zero message if the channel is not recognized.
-        """
-        try:
-            if input_channel in self.input_channel_types:
-                # Create a DataEntry instance with the correct type
-                data_entry: DataEntry[DataModel] = DataEntry(
-                    channel=input_channel, data=input_message.data
-                )
-                await self.queue.put(data_entry)
-            else:
-                logger.warning(f"Unrecognized input channel: {input_channel}")
-                yield input_channel, Message[Text](data=Text(text=""))
-        except Exception as e:
-            logger.error(f"Error handling event: {e}")

From 96f8b6cee73338a7db106b26c507a3f698956368 Mon Sep 17 00:00:00 2001
From: "autofix-ci[bot]" <114827586+autofix-ci[bot]@users.noreply.github.com>
Date: Wed, 13 Nov 2024 01:08:18 +0000
Subject: [PATCH 06/15] [autofix.ci] apply automated fixes

---
 sotopia/experimental/agents/base_agent.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/sotopia/experimental/agents/base_agent.py b/sotopia/experimental/agents/base_agent.py
index 4202f740e..902ad1320 100644
--- a/sotopia/experimental/agents/base_agent.py
+++ b/sotopia/experimental/agents/base_agent.py
@@ -113,7 +113,7 @@ async def send(self, action: T_agent_action) -> None:
         for output_channel, output_channel_type in self.output_channel_types.items():
             await self.r.publish(
                 output_channel,
-                Message[output_channel_type](data=action).model_dump_json(), # type:ignore[valid-type] 
+                Message[output_channel_type](data=action).model_dump_json(),  # type:ignore[valid-type]
             )
 
     async def _task_scheduler(self) -> None:

From d72e21ecf03a6411316f95fb43566cf88dd67b2f Mon Sep 17 00:00:00 2001
From: Arpandeep Khatua <arpandeepk@gmail.com>
Date: Tue, 12 Nov 2024 17:11:39 -0800
Subject: [PATCH 07/15] Correcting mypy error in tick agent

---
 .../experimental/tick_and_echo_agents/tick_and_echo_agents.py   | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/examples/experimental/tick_and_echo_agents/tick_and_echo_agents.py b/examples/experimental/tick_and_echo_agents/tick_and_echo_agents.py
index d05feb218..7cb65ddd9 100644
--- a/examples/experimental/tick_and_echo_agents/tick_and_echo_agents.py
+++ b/examples/experimental/tick_and_echo_agents/tick_and_echo_agents.py
@@ -1,6 +1,6 @@
 from aact import NodeFactory
 from aact.messages import Text, Tick
-from sotopia.experimental.agents import BaseAgent
+from sotopia.experimental.agents.base_agent import BaseAgent
 
 
 @NodeFactory.register("simple_echo_agent")

From b5facf237de864661e8cb967e732d194f37227f6 Mon Sep 17 00:00:00 2001
From: Arpandeep Khatua <arpandeepk@gmail.com>
Date: Fri, 15 Nov 2024 08:28:14 -0800
Subject: [PATCH 08/15] Moving everything to examples since its not specific to
 sotopia

---
 .../experimental/llm_agent}/llm_agent.py      | 61 ++++++++++++++++++-
 .../experimental/llm_agent/llm_agent.toml     | 10 +--
 sotopia/experimental/agents/base_agent.py     | 56 +----------------
 3 files changed, 64 insertions(+), 63 deletions(-)
 rename {sotopia/experimental/agents => examples/experimental/llm_agent}/llm_agent.py (90%)

diff --git a/sotopia/experimental/agents/llm_agent.py b/examples/experimental/llm_agent/llm_agent.py
similarity index 90%
rename from sotopia/experimental/agents/llm_agent.py
rename to examples/experimental/llm_agent/llm_agent.py
index 789deb8a6..f8f734a44 100644
--- a/sotopia/experimental/agents/llm_agent.py
+++ b/examples/experimental/llm_agent/llm_agent.py
@@ -1,11 +1,16 @@
 import logging
 import sys
-
+from enum import Enum
 from rich.logging import RichHandler
+from pydantic import Field
+
+from typing import Optional
 
 from aact import Message, NodeFactory
-from aact.messages import Text, Tick
-from sotopia.experimental.agents.base_agent import BaseAgent, AgentAction, ActionType
+from aact.messages import Text, Tick, DataModel
+from aact.messages.registry import DataModelFactory
+
+from sotopia.experimental.agents.base_agent import BaseAgent
 
 from sotopia.generation_utils import agenerate
 from sotopia.generation_utils.generate import StrOutputParser
@@ -27,6 +32,56 @@
     handlers=[RichHandler()],
 )
 
+class ActionType(Enum):
+    NONE = "none"
+    SPEAK = "speak"
+    NON_VERBAL = "non-verbal"
+    LEAVE = "leave"
+    THOUGHT = "thought"
+    BROWSE = "browse"
+    BROWSE_ACTION = "browse_action"
+    READ = "read"
+    WRITE = "write"
+    RUN = "run"
+
+    def __str__(self) -> str:
+        return self.value
+
+    def __eq__(self, other: object) -> bool:
+        if isinstance(other, ActionType):
+            return self.value == other.value
+        elif isinstance(other, str):
+            return self.value == other
+        else:
+            return NotImplemented
+
+
+@DataModelFactory.register("agent_action")
+class AgentAction(DataModel):
+    agent_name: str = Field(description="the name of the agent")
+    action_type: ActionType = Field(
+        description="whether to speak at this turn or choose to not do anything"
+    )
+    argument: str = Field(
+        description="the utterance if choose to speak, the expression or gesture if choose non-verbal communication, or the physical action if choose action"
+    )
+    path: Optional[str] = Field(description="path of file")
+
+    def to_natural_language(self) -> str:
+        action_descriptions = {
+            ActionType.NONE: "did nothing",
+            ActionType.SPEAK: f'said: "{self.argument}"',
+            ActionType.THOUGHT: f'thought: "{self.argument}"',
+            ActionType.BROWSE: f'browsed: "{self.argument}"',
+            ActionType.RUN: f'ran: "{self.argument}"',
+            ActionType.READ: f'read: "{self.argument}"',
+            ActionType.WRITE: f'wrote: "{self.argument}"',
+            ActionType.NON_VERBAL: f"[{self.action_type.value}] {self.argument}",
+            ActionType.LEAVE: "left the conversation",
+        }
+
+        return action_descriptions.get(self.action_type, "performed an unknown action")
+
 
 @NodeFactory.register("llm_agent")
 class LLMAgent(BaseAgent[AgentAction | Tick | Text, AgentAction]):
diff --git a/examples/experimental/llm_agent/llm_agent.toml b/examples/experimental/llm_agent/llm_agent.toml
index 63061a798..6878cb8d4 100644
--- a/examples/experimental/llm_agent/llm_agent.toml
+++ b/examples/experimental/llm_agent/llm_agent.toml
@@ -1,5 +1,5 @@
 redis_url = "redis://localhost:6379/0"
-extra_modules = ["sotopia.experimental.agents.llm_agent"]
+extra_modules = ["sotopia.examples.experimental.llm_agent.llm_agent"]
 
 
 [[nodes]]
@@ -9,8 +9,8 @@ node_class = "llm_agent"
 [nodes.node_args]
 query_interval = 5
 output_channel = "Jack:Jane"
-input_text_channels = ["Jane:Jack", "Runtime:Agent"]
-input_env_channels = []
+input_text_channels = ["Jane:Jack"]
+input_env_channels = ["Runtime:Agent"]
 input_tick_channel = "tick/secs/1"
 goal = "Your goal is to effectively test Jane's technical ability and finally decide if she has passed the interview. Make sure to also evaluate her communication skills, problem-solving approach, and enthusiasm."
 model_name = "gpt-4o-mini"
@@ -23,8 +23,8 @@ node_class = "llm_agent"
 [nodes.node_args]
 query_interval = 7
 output_channel = "Jane:Jack"
-input_text_channels = ["Jack:Jane", "Runtime:Agent"]
-input_env_channels = []
+input_text_channels = ["Jack:Jane"]
+input_env_channels = ["Runtime:Agent"]
 input_tick_channel = "tick/secs/1"
 goal = "Your goal is to do well in the interview by demonstrating your technical skills, clear communication, and enthusiasm for the position. Stay calm, ask clarifying questions when needed, and confidently explain your thought process."
 model_name = "gpt-4o-mini"
diff --git a/sotopia/experimental/agents/base_agent.py b/sotopia/experimental/agents/base_agent.py
index 902ad1320..a7bbafae6 100644
--- a/sotopia/experimental/agents/base_agent.py
+++ b/sotopia/experimental/agents/base_agent.py
@@ -1,75 +1,21 @@
 import asyncio
 import sys
 
-from enum import Enum
-from pydantic import Field
 
 if sys.version_info < (3, 11):
     from typing_extensions import Self
 else:
     from typing import Self
 
-from typing import Any, AsyncIterator, TypeVar, Optional
+from typing import Any, AsyncIterator, TypeVar
 
 from aact import Message, Node
 from aact.messages import DataModel
-from aact.messages.registry import DataModelFactory
 
 T_agent_observation = TypeVar("T_agent_observation", bound=DataModel)
 T_agent_action = TypeVar("T_agent_action", bound=DataModel)
 
 
-class ActionType(Enum):
-    NONE = "none"
-    SPEAK = "speak"
-    NON_VERBAL = "non-verbal"
-    LEAVE = "leave"
-    THOUGHT = "thought"
-    BROWSE = "browse"
-    BROWSE_ACTION = "browse_action"
-    READ = "read"
-    WRITE = "write"
-    RUN = "run"
-
-    def __str__(self) -> str:
-        return self.value
-
-    def __eq__(self, other: object) -> bool:
-        if isinstance(other, ActionType):
-            return self.value == other.value
-        elif isinstance(other, str):
-            return self.value == other
-        else:
-            return NotImplemented
-
-
-@DataModelFactory.register("agent_action")
-class AgentAction(DataModel):
-    agent_name: str = Field(description="the name of the agent")
-    action_type: ActionType = Field(
-        description="whether to speak at this turn or choose to not do anything"
-    )
-    argument: str = Field(
-        description="the utterance if choose to speak, the expression or gesture if choose non-verbal communication, or the physical action if choose action"
-    )
-    path: Optional[str] = Field(description="path of file")
-
-    def to_natural_language(self) -> str:
-        action_descriptions = {
-            ActionType.NONE: "did nothing",
-            ActionType.SPEAK: f'said: "{self.argument}"',
-            ActionType.THOUGHT: f'thought: "{self.argument}"',
-            ActionType.BROWSE: f'browsed: "{self.argument}"',
-            ActionType.RUN: f'ran: "{self.argument}"',
-            ActionType.READ: f'read: "{self.argument}"',
-            ActionType.WRITE: f'wrote: "{self.argument}"',
-            ActionType.NON_VERBAL: f"[{self.action_type.value}] {self.argument}",
-            ActionType.LEAVE: "left the conversation",
-        }
-
-        return action_descriptions.get(self.action_type, "performed an unknown action")
-
-
 class BaseAgent(Node[T_agent_observation, T_agent_action]):
     def __init__(
         self,

From 558e582433f3892bea94342d816fa43db8781e93 Mon Sep 17 00:00:00 2001
From: "autofix-ci[bot]" <114827586+autofix-ci[bot]@users.noreply.github.com>
Date: Fri, 15 Nov 2024 16:29:04 +0000
Subject: [PATCH 09/15] [autofix.ci] apply automated fixes

---
 examples/experimental/llm_agent/llm_agent.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/examples/experimental/llm_agent/llm_agent.py b/examples/experimental/llm_agent/llm_agent.py
index f8f734a44..d6af3aaca 100644
--- a/examples/experimental/llm_agent/llm_agent.py
+++ b/examples/experimental/llm_agent/llm_agent.py
@@ -32,6 +32,7 @@
     handlers=[RichHandler()],
 )
 
+
 class ActionType(Enum):
     NONE = "none"
     SPEAK = "speak"

From f687c496436fcb230717595b020303c07b897600 Mon Sep 17 00:00:00 2001
From: Arpandeep Khatua <arpandeepk@gmail.com>
Date: Fri, 15 Nov 2024 08:36:25 -0800
Subject: [PATCH 10/15] Adding some additional final checks

---
 examples/experimental/llm_agent/llm_agent.py  | 37 ++++++++++---------
 .../experimental/llm_agent/llm_agent.toml     |  2 +-
 2 files changed, 21 insertions(+), 18 deletions(-)

diff --git a/examples/experimental/llm_agent/llm_agent.py b/examples/experimental/llm_agent/llm_agent.py
index d6af3aaca..67b0025f3 100644
--- a/examples/experimental/llm_agent/llm_agent.py
+++ b/examples/experimental/llm_agent/llm_agent.py
@@ -310,23 +310,26 @@ async def aact(self, message: AgentAction | Tick | Text) -> AgentAction:
             case Tick():
                 self.count_ticks += 1
                 if self.count_ticks % self.query_interval == 0:
-                    template = self.get_action_template(
-                        [action for action in ActionType]
-                    )
-
-                    agent_action = await agenerate(
-                        model_name=self.model_name,
-                        template=template,
-                        input_values={
-                            "message_history": self._format_message_history(
-                                self.message_history
-                            ),
-                            "goal": self.goal,
-                            "agent_name": self.name,
-                        },
-                        temperature=0.7,
-                        output_parser=StrOutputParser(),
-                    )
+                    try:
+                        template = self.get_action_template(
+                            [action for action in ActionType]
+                        )
+
+                        agent_action = await agenerate(
+                            model_name=self.model_name,
+                            template=template,
+                            input_values={
+                                "message_history": self._format_message_history(
+                                    self.message_history
+                                ),
+                                "goal": self.goal,
+                                "agent_name": self.name,
+                            },
+                            temperature=0.7,
+                            output_parser=StrOutputParser(),
+                        )
+                    except Exception as e:
+                        print(f"Error during agenerate: {e}")
 
                     agent_action = (
                         agent_action.replace("```", "")
diff --git a/examples/experimental/llm_agent/llm_agent.toml b/examples/experimental/llm_agent/llm_agent.toml
index 6878cb8d4..daa1b3d4b 100644
--- a/examples/experimental/llm_agent/llm_agent.toml
+++ b/examples/experimental/llm_agent/llm_agent.toml
@@ -1,5 +1,5 @@
 redis_url = "redis://localhost:6379/0"
-extra_modules = ["sotopia.examples.experimental.llm_agent.llm_agent"]
+extra_modules = ["examples.experimental.llm_agent.llm_agent"]
 
 
 [[nodes]]

From 3a9e8a0800239bafdbf8420b257cdc8514eaa806 Mon Sep 17 00:00:00 2001
From: Arpandeep Khatua <arpandeepk@gmail.com>
Date: Fri, 15 Nov 2024 10:42:20 -0800
Subject: [PATCH 11/15] Renaming to interview openhands

---
 .../interview_openhands.toml}                                   | 2 +-
 .../{llm_agent => interview_openhands}/llm_agent.py             | 0
 2 files changed, 1 insertion(+), 1 deletion(-)
 rename examples/experimental/{llm_agent/llm_agent.toml => interview_openhands/interview_openhands.toml} (94%)
 rename examples/experimental/{llm_agent => interview_openhands}/llm_agent.py (100%)

diff --git a/examples/experimental/llm_agent/llm_agent.toml b/examples/experimental/interview_openhands/interview_openhands.toml
similarity index 94%
rename from examples/experimental/llm_agent/llm_agent.toml
rename to examples/experimental/interview_openhands/interview_openhands.toml
index daa1b3d4b..805d4112e 100644
--- a/examples/experimental/llm_agent/llm_agent.toml
+++ b/examples/experimental/interview_openhands/interview_openhands.toml
@@ -1,5 +1,5 @@
 redis_url = "redis://localhost:6379/0"
-extra_modules = ["examples.experimental.llm_agent.llm_agent"]
+extra_modules = ["examples.experimental.interview_openhands.llm_agent"]
 
 
 [[nodes]]
diff --git a/examples/experimental/llm_agent/llm_agent.py b/examples/experimental/interview_openhands/llm_agent.py
similarity index 100%
rename from examples/experimental/llm_agent/llm_agent.py
rename to examples/experimental/interview_openhands/llm_agent.py

From 7d514160f7ae6f7fede1379d0321df8f1dec434e Mon Sep 17 00:00:00 2001
From: Arpandeep Khatua <arpandeepk@gmail.com>
Date: Sat, 16 Nov 2024 10:34:28 -0800
Subject: [PATCH 12/15] Adding documentation

---
 .../interview_openhands/readme.md             | 189 ++++++++++++++++++
 1 file changed, 189 insertions(+)
 create mode 100644 examples/experimental/interview_openhands/readme.md

diff --git a/examples/experimental/interview_openhands/readme.md b/examples/experimental/interview_openhands/readme.md
new file mode 100644
index 000000000..82fc2533e
--- /dev/null
+++ b/examples/experimental/interview_openhands/readme.md
@@ -0,0 +1,189 @@
+# OpenHands Interview Example
+
+In this example, we explore the interaction between two LLM agents, "Jack" and "Jane", within an OpenHands runtime node. These agents are designed to simulate an interview scenario where "Jack" evaluates "Jane's" technical abilities and communication skills.
+
+## Agents Overview
+
+- **Jack (Interviewer)**: An LLM agent tasked with assessing Jane's technical skills, communication, problem-solving approach, and enthusiasm.
+- **Jane (Interviewee)**: An LLM agent aiming to perform well in the interview by showcasing her skills and enthusiasm.
+
+## Customizing the Scenario
+
+You can modify the `interview_openhands.toml` file to simulate different scenarios by changing the goals, model names, and communication channels. This flexibility allows you to tailor the interaction to various use cases, such as customer service simulations, collaborative problem-solving, or educational tutoring.
+
+## Actions Available to LLM Agents
+
+- **speak**: Communicate with other agents by sending a message.
+- **thought**: Make a plan or set a goal internally.
+- **none**: Choose not to take any action, often used when waiting for data.
+- **non-verbal**: Perform a non-verbal action, such as a gesture.
+- **browse**: Open a web page.
+- **browse_action**: Perform actions on a web browser, such as navigating or interacting with elements.
+- **read**: Read the content of a file.
+- **write**: Write content to a file.
+- **run**: Execute a command in a Linux shell.
+- **leave**: Exit the conversation when goals are completed or abandoned.
+
+These actions enable the agents to interact dynamically with their environment and each other, providing a robust framework for simulating complex scenarios.
+
+## Implementation of the Agents
+
+The implementation of these agents can be found in `llm_agent.py`. The agents are configured to interact with each other through specified channels, simulating a real-time interview process.
+
+## Dataflow Configuration
+
+The dataflow file `interview_openhands.toml` is used to configure the interaction between the agents. It acts as a blueprint, defining how the agents communicate and process information.
+
+### Example Configuration
+
+```toml
+[[nodes]]
+node_name = "Jack"
+node_class = "llm_agent"
+
+[nodes.node_args]
+query_interval = 5
+output_channel = "Jack:Jane"
+input_text_channels = ["Jane:Jack"]
+input_env_channels = ["Runtime:Agent"]
+input_tick_channel = "tick/secs/1"
+goal = "Your goal is to effectively test Jane's technical ability and finally decide if she has passed the interview. Make sure to also evaluate her communication skills, problem-solving approach, and enthusiasm."
+model_name = "gpt-4o-mini"
+agent_name = "Jack"
+
+[[nodes]]
+node_name = "Jane"
+node_class = "llm_agent"
+
+[nodes.node_args]
+query_interval = 7
+output_channel = "Jane:Jack"
+input_text_channels = ["Jack:Jane"]
+input_env_channels = ["Runtime:Agent"]
+input_tick_channel = "tick/secs/1"
+goal = "Your goal is to do well in the interview by demonstrating your technical skills, clear communication, and enthusiasm for the position. Stay calm, ask clarifying questions when needed, and confidently explain your thought process."
+model_name = "gpt-4o-mini"
+agent_name = "Jane"
+```
+
+## Auxiliary Nodes
+
+In addition to the agents, there are auxiliary nodes that facilitate the interaction:
+
+- **Tick Node**: Provides a stable clock signal for synchronization.
+- **Print Node**: Outputs real-time messages for monitoring the interaction.
+
+### Auxiliary Node Configuration
+
+```toml
+[[nodes]]
+node_name = "tick"
+node_class = "tick"
+
+[[nodes]]
+node_name = "print"
+node_class = "print"
+
+[nodes.node_args.print_channel_types]
+"tick/secs/1" = "tick"
+"Jane:Jack" = "agent_action"
+"Jack:Jane" = "agent_action"
+```
+
+## Running the Example
+
+### Prerequisites
+
+1. **Redis**: Ensure Redis is hosted at `localhost:6379`. Update the `redis_url` in the `interview_openhands.toml` file if using a different Redis database.
+
+2. **Setting Up OpenHands Runtime**: Clone the repository and set up the environment.
+
+   - **Clone the Repository**: Clone the repository from GitHub to get the necessary files and configurations.
+     ```bash
+     git clone https://github.com/akhatua2/aact-openhands
+     cd aact-openhands
+     ```
+
+   - **Python and Poetry**: Ensure you have Python 3.12 installed. Install Poetry for dependency management if you haven't already.
+     ```bash
+     curl -sSL https://install.python-poetry.org | python3 -
+     ```
+
+   - **Install Dependencies**: Navigate to the project directory and install the dependencies using Poetry.
+     ```bash
+     poetry install
+     ```
+
+   - **Environment Variables**: Set up your environment variables. Copy the `.env.example` file to a new file named `.env` in the root of your project directory and replace the placeholder values with your actual API keys and URLs.
+     ```bash
+     cp .env.example .env
+     ```
+
+     Edit the `.env` file:
+     ```plaintext
+     MODAL_API_TOKEN_ID=your_actual_modal_api_token_id
+     MODAL_API_TOKEN_SECRET=your_actual_modal_api_token_secret
+     ```
+   - **Running the OpenHands Node**
+
+        To run the OpenHands node with the provided configuration, use the following command:
+
+        ```bash
+        poetry run aact run-dataflow examples/openhands_node.toml
+        ```
+
+        Upon successful execution, you should see output similar to the following:
+
+        ```bash
+        16:41:26 - openhands:INFO: openhands_node.py:120 - --------------------
+        16:41:26 - openhands:INFO: openhands_node.py:121 - RUNTIME CONNECTED
+        16:41:26 - openhands:INFO: openhands_node.py:122 - --------------------
+        16:41:26 - openhands:INFO: openhands_node.py:127 - Runtime initialization took 157.77 seconds.
+        ```
+
+These steps ensure that you have the necessary environment and dependencies to run the OpenHands node and the example scenario.
+
+### Command
+
+To run this interview setting example, use the following command:
+
+```bash
+uv run aact run-dataflow examples/experimental/interview_openhands/interview_openhands.toml
+```
+
+## Expected Output
+
+You will see JSON strings printed out, representing messages exchanged between the nodes. These messages include timestamps for easy debugging and data recording, demonstrating the real-time interaction between the agents.
+
+```json
+{"timestamp":"2024-11-12T17:27:27.797795","channel":"tick/secs/1","data":{"data_type":"tick","tick":0}}
+{"timestamp":"2024-11-12T17:27:28.810269","channel":"tick/secs/1","data":{"data_type":"tick","tick":1}}
+{"timestamp":"2024-11-12T17:27:29.805630","channel":"tick/secs/1","data":{"data_type":"tick","tick":2}}
+{"timestamp":"2024-11-12T17:27:30.803932","channel":"tick/secs/1","data":{"data_type":"tick","tick":3}}
+{"timestamp":"2024-11-12T17:27:31.804342","channel":"tick/secs/1","data":{"data_type":"tick","tick":4}}
+{"timestamp":"2024-11-12T17:27:32.805544","channel":"tick/secs/1","data":{"data_type":"tick","tick":5}}
+{"timestamp":"2024-11-12T17:27:33.094975","channel":"Jack:Jane","data":{"data_type":"agent_action","agent_name":"Jack","
+action_type":"speak","argument":"Hi Jane! I'm Jack, a Principal Software Engineer here at Fintech. It's great to meet you! Could you please introduce yourself and tell me a little about your education and any relevant experiences you've had?","path":""}}                                                                                                          
+{"timestamp":"2024-11-12T17:27:33.801111","channel":"tick/secs/1","data":{"data_type":"tick","tick":6}}
+{"timestamp":"2024-11-12T17:27:34.803471","channel":"tick/secs/1","data":{"data_type":"tick","tick":7}}
+{"timestamp":"2024-11-12T17:27:35.118890","channel":"Jane:Jack","data":{"data_type":"agent_action","agent_name":"Jane","
+action_type":"speak","argument":"Hi Jack! I'm Jane, a college senior at Stanford University majoring in Computer Science. I've had a strong focus on software development, especially in fintech applications. I recently completed a project where I built a budgeting app using React and Node.js, which helped users manage their finances effectively. I'm excited about the opportunity to bring my skills to your team.","path":""}}                                                      
+{"timestamp":"2024-11-12T17:27:35.803424","channel":"tick/secs/1","data":{"data_type":"tick","tick":8}}
+{"timestamp":"2024-11-12T17:27:36.800113","channel":"tick/secs/1","data":{"data_type":"tick","tick":9}}
+{"timestamp":"2024-11-12T17:27:37.732772","channel":"Jack:Jane","data":{"data_type":"agent_action","agent_name":"Jack","
+action_type":"speak","argument":"That's a fantastic introduction, Jane! Your project sounds really interesting. Now, let's dive into some technical questions. Are you ready?","path":""}}                                                      
+{"timestamp":"2024-11-12T17:27:37.802020","channel":"tick/secs/1","data":{"data_type":"tick","tick":10}}
+{"timestamp":"2024-11-12T17:27:38.800835","channel":"tick/secs/1","data":{"data_type":"tick","tick":11}}
+{"timestamp":"2024-11-12T17:27:39.804363","channel":"tick/secs/1","data":{"data_type":"tick","tick":12}}
+```
+
+This output shows the following events:
+
+1. The tick node sends a tick message, indicating the passage of time.
+2. "Jack" initiates the conversation by introducing himself and asking "Jane" to introduce herself.
+3. "Jane" responds with her introduction, highlighting her education and relevant experiences.
+4. "Jack" acknowledges Jane's introduction and prepares to ask technical questions.
+
+Each message includes a timestamp for easy debugging and data recording. The interaction demonstrates the agents' ability to simulate a realistic interview scenario, with messages exchanged in real-time.
+
+

From 8e0d9da0a788355225d45f1f08c37971c4478309 Mon Sep 17 00:00:00 2001
From: "autofix-ci[bot]" <114827586+autofix-ci[bot]@users.noreply.github.com>
Date: Sat, 16 Nov 2024 18:35:19 +0000
Subject: [PATCH 13/15] [autofix.ci] apply automated fixes

---
 examples/experimental/interview_openhands/readme.md | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/examples/experimental/interview_openhands/readme.md b/examples/experimental/interview_openhands/readme.md
index 82fc2533e..07552cc9e 100644
--- a/examples/experimental/interview_openhands/readme.md
+++ b/examples/experimental/interview_openhands/readme.md
@@ -163,15 +163,15 @@ You will see JSON strings printed out, representing messages exchanged between t
 {"timestamp":"2024-11-12T17:27:31.804342","channel":"tick/secs/1","data":{"data_type":"tick","tick":4}}
 {"timestamp":"2024-11-12T17:27:32.805544","channel":"tick/secs/1","data":{"data_type":"tick","tick":5}}
 {"timestamp":"2024-11-12T17:27:33.094975","channel":"Jack:Jane","data":{"data_type":"agent_action","agent_name":"Jack","
-action_type":"speak","argument":"Hi Jane! I'm Jack, a Principal Software Engineer here at Fintech. It's great to meet you! Could you please introduce yourself and tell me a little about your education and any relevant experiences you've had?","path":""}}                                                                                                          
+action_type":"speak","argument":"Hi Jane! I'm Jack, a Principal Software Engineer here at Fintech. It's great to meet you! Could you please introduce yourself and tell me a little about your education and any relevant experiences you've had?","path":""}}
 {"timestamp":"2024-11-12T17:27:33.801111","channel":"tick/secs/1","data":{"data_type":"tick","tick":6}}
 {"timestamp":"2024-11-12T17:27:34.803471","channel":"tick/secs/1","data":{"data_type":"tick","tick":7}}
 {"timestamp":"2024-11-12T17:27:35.118890","channel":"Jane:Jack","data":{"data_type":"agent_action","agent_name":"Jane","
-action_type":"speak","argument":"Hi Jack! I'm Jane, a college senior at Stanford University majoring in Computer Science. I've had a strong focus on software development, especially in fintech applications. I recently completed a project where I built a budgeting app using React and Node.js, which helped users manage their finances effectively. I'm excited about the opportunity to bring my skills to your team.","path":""}}                                                      
+action_type":"speak","argument":"Hi Jack! I'm Jane, a college senior at Stanford University majoring in Computer Science. I've had a strong focus on software development, especially in fintech applications. I recently completed a project where I built a budgeting app using React and Node.js, which helped users manage their finances effectively. I'm excited about the opportunity to bring my skills to your team.","path":""}}
 {"timestamp":"2024-11-12T17:27:35.803424","channel":"tick/secs/1","data":{"data_type":"tick","tick":8}}
 {"timestamp":"2024-11-12T17:27:36.800113","channel":"tick/secs/1","data":{"data_type":"tick","tick":9}}
 {"timestamp":"2024-11-12T17:27:37.732772","channel":"Jack:Jane","data":{"data_type":"agent_action","agent_name":"Jack","
-action_type":"speak","argument":"That's a fantastic introduction, Jane! Your project sounds really interesting. Now, let's dive into some technical questions. Are you ready?","path":""}}                                                      
+action_type":"speak","argument":"That's a fantastic introduction, Jane! Your project sounds really interesting. Now, let's dive into some technical questions. Are you ready?","path":""}}
 {"timestamp":"2024-11-12T17:27:37.802020","channel":"tick/secs/1","data":{"data_type":"tick","tick":10}}
 {"timestamp":"2024-11-12T17:27:38.800835","channel":"tick/secs/1","data":{"data_type":"tick","tick":11}}
 {"timestamp":"2024-11-12T17:27:39.804363","channel":"tick/secs/1","data":{"data_type":"tick","tick":12}}
@@ -185,5 +185,3 @@ This output shows the following events:
 4. "Jack" acknowledges Jane's introduction and prepares to ask technical questions.
 
 Each message includes a timestamp for easy debugging and data recording. The interaction demonstrates the agents' ability to simulate a realistic interview scenario, with messages exchanged in real-time.
-
-

From fa65167cf26a22e339d4de3fb26e91150da0f08c Mon Sep 17 00:00:00 2001
From: Arpandeep Khatua <arpandeepk@gmail.com>
Date: Sat, 16 Nov 2024 10:39:37 -0800
Subject: [PATCH 14/15] Updating the readme

---
 .../experimental/interview_openhands/readme.md   | 16 ++++++----------
 1 file changed, 6 insertions(+), 10 deletions(-)

diff --git a/examples/experimental/interview_openhands/readme.md b/examples/experimental/interview_openhands/readme.md
index 07552cc9e..81edcf11b 100644
--- a/examples/experimental/interview_openhands/readme.md
+++ b/examples/experimental/interview_openhands/readme.md
@@ -7,11 +7,11 @@ In this example, we explore the interaction between two LLM agents, "Jack" and "
 - **Jack (Interviewer)**: An LLM agent tasked with assessing Jane's technical skills, communication, problem-solving approach, and enthusiasm.
 - **Jane (Interviewee)**: An LLM agent aiming to perform well in the interview by showcasing her skills and enthusiasm.
 
-## Customizing the Scenario
+### Customizing the Scenario
 
 You can modify the `interview_openhands.toml` file to simulate different scenarios by changing the goals, model names, and communication channels. This flexibility allows you to tailor the interaction to various use cases, such as customer service simulations, collaborative problem-solving, or educational tutoring.
 
-## Actions Available to LLM Agents
+### Actions Available to LLM Agents
 
 - **speak**: Communicate with other agents by sending a message.
 - **thought**: Make a plan or set a goal internally.
@@ -26,10 +26,6 @@ You can modify the `interview_openhands.toml` file to simulate different scenari
 
 These actions enable the agents to interact dynamically with their environment and each other, providing a robust framework for simulating complex scenarios.
 
-## Implementation of the Agents
-
-The implementation of these agents can be found in `llm_agent.py`. The agents are configured to interact with each other through specified channels, simulating a real-time interview process.
-
 ## Dataflow Configuration
 
 The dataflow file `interview_openhands.toml` is used to configure the interaction between the agents. It acts as a blueprint, defining how the agents communicate and process information.
@@ -66,14 +62,14 @@ model_name = "gpt-4o-mini"
 agent_name = "Jane"
 ```
 
-## Auxiliary Nodes
+### Auxiliary Nodes
 
 In addition to the agents, there are auxiliary nodes that facilitate the interaction:
 
 - **Tick Node**: Provides a stable clock signal for synchronization.
 - **Print Node**: Outputs real-time messages for monitoring the interaction.
 
-### Auxiliary Node Configuration
+#### Auxiliary Node Configuration
 
 ```toml
 [[nodes]]
@@ -151,7 +147,7 @@ To run this interview setting example, use the following command:
 uv run aact run-dataflow examples/experimental/interview_openhands/interview_openhands.toml
 ```
 
-## Expected Output
+### Expected Output
 
 You will see JSON strings printed out, representing messages exchanged between the nodes. These messages include timestamps for easy debugging and data recording, demonstrating the real-time interaction between the agents.
 
@@ -184,4 +180,4 @@ This output shows the following events:
 3. "Jane" responds with her introduction, highlighting her education and relevant experiences.
 4. "Jack" acknowledges Jane's introduction and prepares to ask technical questions.
 
-Each message includes a timestamp for easy debugging and data recording. The interaction demonstrates the agents' ability to simulate a realistic interview scenario, with messages exchanged in real-time.
+Each message includes a timestamp for easy debugging and data recording. The interaction demonstrates the agents' ability to simulate a realistic interview scenario, with messages exchanged in real-time.
\ No newline at end of file

From c2e3656dc57664ce8bda91cc30275f6ee0950dbb Mon Sep 17 00:00:00 2001
From: "autofix-ci[bot]" <114827586+autofix-ci[bot]@users.noreply.github.com>
Date: Sat, 16 Nov 2024 18:40:18 +0000
Subject: [PATCH 15/15] [autofix.ci] apply automated fixes

---
 examples/experimental/interview_openhands/readme.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/examples/experimental/interview_openhands/readme.md b/examples/experimental/interview_openhands/readme.md
index 81edcf11b..fc78eb25d 100644
--- a/examples/experimental/interview_openhands/readme.md
+++ b/examples/experimental/interview_openhands/readme.md
@@ -180,4 +180,4 @@ This output shows the following events:
 3. "Jane" responds with her introduction, highlighting her education and relevant experiences.
 4. "Jack" acknowledges Jane's introduction and prepares to ask technical questions.
 
-Each message includes a timestamp for easy debugging and data recording. The interaction demonstrates the agents' ability to simulate a realistic interview scenario, with messages exchanged in real-time.
\ No newline at end of file
+Each message includes a timestamp for easy debugging and data recording. The interaction demonstrates the agents' ability to simulate a realistic interview scenario, with messages exchanged in real-time.