I’ve been fascinated by debates—the strategic framing, the sharp retorts, and the fastidiously timed comebacks. Debates aren’t simply entertaining; they’re structured battles of concepts, pushed by logic and proof. Just lately, I began questioning: might we replicate that dynamic utilizing AI brokers—having them debate one another autonomously, full with real-time fact-checking and moderation? The consequence was Deb8flow, an autonomous AI debating setting powered by LangGraph, OpenAI’s GPT-4o mannequin, and the brand new built-in Internet Search function.
In Deb8flow, two brokers—Professional and Con—sq. off on a given subject whereas a Moderator manages turn-taking. A devoted Truth Checker critiques each declare in actual time utilizing GPT-4o’s new shopping capabilities, and a last Choose evaluates the arguments for high quality and coherence. If an agent repeatedly makes factual errors, they’re routinely disqualified—guaranteeing the controversy stays grounded in reality.
This text gives an in-depth take a look at the superior structure and dynamic workflows that energy autonomous AI debates. I’ll stroll you thru how Deb8flow’s modular design leverages LangGraph’s state administration and conditional routing, alongside GPT-4o’s capabilities.
Even should you’re new to AI brokers or LangGraph (see sources [1] and [2] for primers), I’ll clarify the important thing ideas clearly. And should you’d prefer to discover additional, the complete mission is on the market on GitHub: iason-solomos/Deb8flow.
Able to see how AI brokers can debate autonomously in follow?
Let’s dive in.
Excessive-Stage Overview: Autonomous Debates with A number of Brokers
In Deb8flow, we orchestrate a formal debate between two AI brokers – one arguing Professional and one Con – full with a Moderator, a Truth Checker, and a last Choose. The talk unfolds autonomously, with every agent taking part in a job in a structured format.
At its core, Deb8flow is a LangGraph-powered agent system, constructed atop LangChain, utilizing GPT-4o to energy every function—Professional, Con, Choose, and past. We use GPT-4o’s preview mannequin with shopping capabilities to allow real-time fact-checking. In essence, the Professional and Con brokers debate; after every assertion, a fact-checker agent makes use of GPT-4o’s net search to catch any hallucinations or inaccuracies in that assertion in actual time. The talk solely continues as soon as the assertion is verified. The entire course of is coordinated by a LangGraph-defined workflow that ensures correct turn-taking and conditional logic.
Excessive-level debate move graph. Every rectangle is an agent node (Professional/Con debaters, Truth Checker, Choose, and many others.), and diamonds are management nodes (Moderator and a router after fact-checking). Strong arrows denote the conventional development, whereas dashed arrows point out retries if a declare fails fact-check. The Choose node outputs the ultimate verdict, then the workflow ends.
Picture generated by the writer with DALL-E
The talk workflow goes via these levels:
- Matter Technology: A Matter Generator agent produces a nuanced, debatable subject for the session (e.g. “Ought to AI be utilized in classroom training?”).
- Opening: The Professional Argument Agent makes a gap assertion in favor of the subject, kicking off the controversy.
- Rebuttal: The Debate Moderator then provides the ground to the Con Argument agent, who rebuts the Professional’s opening assertion.
- Counter: The Moderator provides the ground again to the Professional agent, who counters the Con agent’s factors.
- Closing: The Moderator switches the ground to the Con agent one final time for a closing argument.
- Judgment: Lastly, the Choose agent critiques the complete debate historical past and evaluates either side primarily based on argument high quality, readability, and persuasiveness. Probably the most convincing aspect wins.
After each single speech, the Truth Checker agent steps in to confirm the factual accuracy of that assertion. If a debater’s declare doesn’t maintain up (e.g. cites a mistaken statistic or “hallucinates” a truth), the workflow triggers a retry: the speaker has to appropriate or modify their assertion. (If both debater accumulates 3 fact-check failures, they’re routinely disqualified for repeatedly spreading inaccuracies, and their opponent wins by default.) This mechanism retains our AI debaters trustworthy and grounded in actuality!
Stipulations and Setup
Earlier than diving into the code, be sure to have the next in place:
- Python 3.12+ put in.
- An OpenAI API key with entry to the GPT-4o mannequin. You may create your individual API key right here: https://platform.openai.com/settings/group/api-keys
- Mission Code: Clone the Deb8flow repository from GitHub (
git clone https://github.com/iason-solomos/Deb8flow.git). The repo features anecessities.txtfor all required packages. Key dependencies embody LangChain/LangGraph (for constructing the agent graph) and the OpenAI Python shopper. - Set up Dependencies: In your mission listing, run:
pip set up -r necessities.txtto put in the required libraries. - Create a
.envfile within the mission root to carry your OpenAI API credentials. It needs to be of the shape:OPENAI_API_KEY_GPT4O = "sk-…" - You can even at any time try the README file: https://github.com/iason-solomos/Deb8flow should you merely need to run the completed app.
As soon as dependencies are put in and the setting variable is ready, you have to be able to run the app. The mission construction is organized for readability:
Deb8flow/
├── configurations/
│ ├── debate_constants.py
│ └── llm_config.py
├── nodes/
│ ├── base_component.py
│ ├── topic_generator_node.py
│ ├── pro_debater_node.py
│ ├── con_debater_node.py
│ ├── debate_moderator_node.py
│ ├── fact_checker_node.py
│ ├── fact_check_router_node.py
│ └── judge_node.py
├── prompts/
│ ├── topic_generator_prompts.py
│ ├── pro_debater_prompts.py
│ ├── con_debater_prompts.py
│ └── … (prompts for different brokers)
├── checks/ (accommodates unit and complete workflow checks)
└── debate_workflow.py
A fast tour of this construction:
configurations/ holds fixed definitions and LLM configuration lessons.
nodes/ accommodates the implementation of every agent or purposeful node within the debate (every of those is a module defining one agent’s habits).
prompts/ shops the immediate templates for the language mannequin (so every agent is aware of learn how to immediate GPT-4o for its particular activity).
debate_workflow.py ties every thing collectively by defining the LangGraph workflow (the graph of nodes and transitions).
debate_state.py defines the shared knowledge construction that the brokers will probably be utilizing on every run.
checks/ contains some fundamental checks and instance runs that can assist you confirm every thing is working.
Beneath the Hood: State Administration and Workflow Setup
To coordinate a posh multi-turn debate, we’d like a shared state and a well-defined move. We’ll begin by taking a look at how Deb8flow defines the debate state and constants, after which see how the LangGraph workflow is constructed.
Defining the Debate State Schema (debate_state.py)
Deb8flow makes use of a shared state (https://langchain-ai.github.io/langgraph/ideas/low_level/#state ) within the type of a Python TypedDict that every one brokers can learn from and replace. This state tracks the controversy’s progress and context – issues like the subject, the historical past of messages, whose flip it’s, and many others. By centralizing this data, every agent node could make selections primarily based on the present state of the controversy.
Hyperlink: debate_state.py
from typing import TypedDict, Record, Dict, Literal
DebateStage = Literal["opening", "rebuttal", "counter", "final_argument"]
class DebateMessage(TypedDict):
speaker: str # e.g. professional or con
content material: str # The message every speaker produced
validated: bool # Whether or not the FactChecker okay’d this message
stage: DebateStage # The stage of the controversy when this message was produced
class DebateState(TypedDict):
debate_topic: str
positions: Dict[str, str]
messages: Record[DebateMessage]
opening_statement_pro_agent: str
stage: str # "opening", "rebuttal", "counter", "final_argument"
speaker: str # "professional" or "con"
times_pro_fact_checked: int # The variety of occasions the professional agent has been fact-checked. If it reaches 3, the professional agent is disqualified.
times_con_fact_checked: int # The variety of occasions the con agent has been fact-checked. If it reaches 3, the con agent is disqualified.Key fields that we have to have within the DebateState embody:
debate_topic(str): The subject being debated.messages(Record[DebateMessage]): An inventory of all messages exchanged up to now. Every message is a dictionary with fields forspeaker(e.g."professional"or"con"or"fact_checker"), the messagecontent material(textual content), avalidatedflag (whether or not it handed fact-check), and thestageof the controversy when it was produced.stage(str): The present debate stage (one in all"opening","rebuttal","counter","final_argument").speaker(str): Whose flip it’s presently ("professional"or"con").times_pro_fact_checked/times_con_fact_checked(int): Counters for what number of occasions all sides has been caught with a false declare. (In our guidelines, if a debater fails fact-check 3 occasions, they could possibly be disqualified or routinely lose.)positions(Dict[str, str]): (Elective) A mapping of every aspect’s basic stance (e.g.,"professional": "In favor of the subject").
By structuring the controversy’s state, brokers discover it straightforward to entry the dialog historical past or verify the present stage, and the management logic can replace the state between turns. The state is actually the reminiscence of the controversy.
Constants and Configuration
To keep away from “magic strings” scattered within the code, we outline some constants in debate_constants.py. For instance, constants for stage names (STAGE_OPENING = "opening", and many others.), speaker identifiers (SPEAKER_PRO = "professional", SPEAKER_CON = "con", and many others.), and node names (NODE_PRO_DEBATER = "pro_debater_node", and many others.). These make the code simpler to keep up and browse.
# Stage names
STAGE_OPENING = "opening"
STAGE_REBUTTAL = "rebuttal"
STAGE_COUNTER = "counter"
STAGE_FINAL_ARGUMENT = "final_argument"
STAGE_END = "finish"
# Audio system
SPEAKER_PRO = "professional"
SPEAKER_CON = "con"
SPEAKER_JUDGE = "choose"
# Node names
NODE_PRO_DEBATER = "pro_debater_node"
NODE_CON_DEBATER = "con_debater_node"
NODE_DEBATE_MODERATOR = "debate_moderator_node"
NODE_JUDGE = "judge_node"
We additionally arrange LLM configuration in llm_config.py. Right here, we outline lessons for OpenAI or Azure OpenAI configs after which create a dictionary llm_config_map mapping mannequin names to their config. For example, we map "gpt-4o" to an OpenAILLMConfig that holds the mannequin title and API key. This manner, each time we have to initialize a GPT-4o agent, we are able to simply do llm_config_map["gpt-4o"] to get the appropriate config. All our fundamental brokers (debaters, subject generator, choose) use this identical GPT-4o configuration.
import os
from dataclasses import dataclass
from typing import Union
@dataclass
class OpenAILLMConfig:
"""
An information class to retailer configuration particulars for OpenAI fashions.
Attributes:
model_name (str): The title of the OpenAI mannequin to make use of.
openai_api_key (str): The API key for authenticating with the OpenAI service.
"""
model_name: str
openai_api_key: str
llm_config_map = {
"gpt-4o": OpenAILLMConfig(
model_name="gpt-4o",
openai_api_key=os.getenv("OPENAI_API_KEY_GPT4O"),
)
}
Constructing the LangGraph Workflow (debate_workflow.py)
With state and configs in place, we assemble the debate workflow graph. LangGraph’s StateGraph is the spine that connects all our agent nodes within the order they need to execute. Right here’s how we set it up:
class DebateWorkflow:
def _initialize_workflow(self) -> StateGraph:
workflow = StateGraph(DebateState)
# Nodes
workflow.add_node("generate_topic_node", GenerateTopicNode(llm_config_map["gpt-4o"]))
workflow.add_node("pro_debater_node", ProDebaterNode(llm_config_map["gpt-4o"]))
workflow.add_node("con_debater_node", ConDebaterNode(llm_config_map["gpt-4o"]))
workflow.add_node("fact_check_node", FactCheckNode())
workflow.add_node("fact_check_router_node", FactCheckRouterNode())
workflow.add_node("debate_moderator_node", DebateModeratorNode())
workflow.add_node("judge_node", JudgeNode(llm_config_map["gpt-4o"]))
# Entry level
workflow.set_entry_point("generate_topic_node")
# Movement
workflow.add_edge("generate_topic_node", "pro_debater_node")
workflow.add_edge("pro_debater_node", "fact_check_node")
workflow.add_edge("con_debater_node", "fact_check_node")
workflow.add_edge("fact_check_node", "fact_check_router_node")
workflow.add_edge("judge_node", END)
return workflow
async def run(self):
workflow = self._initialize_workflow()
graph = workflow.compile()
# graph.get_graph().draw_mermaid_png(output_file_path="workflow_graph.png")
initial_state = {
"subject": "",
"positions": {}
}
final_state = await graph.ainvoke(initial_state, config={"recursion_limit": 50})
return final_state
Let’s break down what’s occurring:
- We initialize a brand new
StateGraphwith ourDebateStatekind because the state schema. - We add every node (agent) to the graph with a reputation. For nodes that want an LLM, we go within the GPT-4o config. For instance,
"pro_debater_node"is added asProDebaterNode(llm_config_map["gpt-4o"]), which means the Professional debater agent will use GPT-4o as its underlying mannequin. - We set the entry level of the graph to
"generate_topic_node". This implies step one of the workflow is to generate a debate subject. - Then we add directed edges to attach nodes. The sides above encode the first sequence: subject -> professional’s flip -> fact-check -> (then a routing determination) -> … finally -> choose -> END. We don’t join the Moderator or Truth Verify Router with static edges, since these nodes use dynamic instructions to redirect the move. The ultimate edge connects the choose to an
ENDmarker to terminate the graph.
When the workflow runs, management will go alongside these edges so as, however each time we hit a router or moderator node, that node will output a command telling the graph which node to go to subsequent (overriding the default edge). That is how we create conditional loops: the fact_check_router_node would possibly ship us again to a debater node for a retry, as an alternative of following a straight line. LangGraph helps this by permitting nodes to return a particular Command object with goto directions.
In abstract, at a excessive degree we’ve outlined an agentic workflow: a graph of autonomous brokers the place management can department and loop primarily based on the brokers’ outputs. Now, let’s discover what every of those agent nodes really does.
Agent Nodes Breakdown
Every stage or function within the debate is encapsulated in a node (agent). In LangGraph, nodes are sometimes easy capabilities, however I wished a extra object-oriented method for readability and reusability. So in Deb8flow, each node is a class with a __call__ technique. All the principle agent lessons inherit from a typical BaseComponent for shared performance. This design makes the system modular: we are able to simply swap out or prolong brokers by modifying their class definitions, and every agent class is liable for its piece of the workflow.
Let’s undergo the important thing brokers one after the other.
BaseComponent – A Reusable Agent Base Class
Most of our agent nodes (just like the debaters and choose) share widespread wants: they use an LLM to generate output, they may have to retry on errors, and they need to observe token utilization. The BaseComponent class (outlined in <a href="https://github.com/iason-solomos/Deb8flow/blob/fundamental/nodes/base_component.py">nodes/base_component.py</a>) offers these widespread options so we don’t repeat code.
class BaseComponent:
"""
A foundational class for managing LLM-based workflows with token monitoring.
Can deal with each Azure OpenAI (AzureChatOpenAI) and OpenAI (ChatOpenAI).
"""
def __init__(
self,
llm_config: Elective[LLMConfig] = None,
temperature: float = 0.0,
max_retries: int = 5,
):
"""
Initializes the BaseComponent with non-compulsory LLM configuration and temperature.
Args:
llm_config (Elective[LLMConfig]): Configuration for both Azure or OpenAI.
temperature (float): Controls the randomness of LLM outputs. Defaults to 0.0.
max_retries (int): What number of occasions to retry on 429 errors.
"""
logger = logging.getLogger(self.__class__.__name__)
tracer = hint.get_tracer(__name__, tracer_provider=get_tracer_provider())
self.logger = logger
self.tracer = tracer
self.llm: Elective[ChatOpenAI] = None
self.output_parser: Elective[StrOutputParser] = None
self.state: Elective[DebateState] = None
self.prompt_template: Elective[ChatPromptTemplate] = None
self.chain: Elective[RunnableSequence] = None
self.paperwork: Elective[List] = None
self.prompt_tokens = 0
self.completion_tokens = 0
self.max_retries = max_retries
if llm_config just isn't None:
self.llm = self._init_llm(llm_config, temperature)
self.output_parser = StrOutputParser()
def _init_llm(self, config: LLMConfig, temperature: float):
"""
Initializes an LLM occasion for both Azure OpenAI or customary OpenAI.
"""
if isinstance(config, AzureOpenAILLMConfig):
# If it is Azure, use the AzureChatOpenAI class
return AzureChatOpenAI(
deployment_name=config.deployment_name,
azure_endpoint=config.azure_endpoint,
openai_api_version=config.openai_api_version,
openai_api_key=config.openai_api_key,
temperature=temperature,
)
elif isinstance(config, OpenAILLMConfig):
# If it is customary OpenAI, use the ChatOpenAI class
return ChatOpenAI(
model_name=config.model_name,
openai_api_key=config.openai_api_key,
temperature=temperature,
)
else:
elevate ValueError("Unsupported LLMConfig kind.")
def validate_initialization(self) -> None:
"""
Ensures we have now an LLM and an output parser.
"""
if not self.llm:
elevate ValueError("LLM just isn't initialized. Guarantee `llm_config` is supplied.")
if not self.output_parser:
elevate ValueError("Output parser just isn't initialized.")
def execute_chain(self, inputs: Any) -> Any:
"""
Executes the LLM chain, tracks token utilization, and retries on 429 errors.
"""
if not self.chain:
elevate ValueError("No chain is initialized for execution.")
retry_wait = 1 # Preliminary wait time in seconds
for try in vary(self.max_retries):
strive:
with get_openai_callback() as cb:
consequence = self.chain.invoke(inputs)
self.logger.information("Immediate Token utilization: %s", cb.prompt_tokens)
self.logger.information("Completion Token utilization: %s", cb.completion_tokens)
self.prompt_tokens = cb.prompt_tokens
self.completion_tokens = cb.completion_tokens
return consequence
besides Exception as e:
# If the error mentions 429, do exponential backoff and retry
if "429" in str(e):
self.logger.warning(
f"Price restrict reached. Retrying in {retry_wait} seconds... "
f"(Try {try + 1}/{self.max_retries})"
)
time.sleep(retry_wait)
retry_wait *= 2
else:
self.logger.error(f"Surprising error: {str(e)}")
elevate e
elevate Exception("API request failed after most variety of retries")
def create_chain(
self, system_template: str, human_template: str
) -> RunnableSequence:
"""
Creates a sequence for unstructured outputs.
"""
self.validate_initialization()
self.prompt_template = ChatPromptTemplate.from_messages(
[
("system", system_template),
("human", human_template),
]
)
self.chain = self.prompt_template | self.llm | self.output_parser
return self.chain
def create_structured_output_chain(
self, system_template: str, human_template: str, output_model: Kind[BaseModel]
) -> RunnableSequence:
"""
Creates a sequence that yields structured outputs (parsed right into a Pydantic mannequin).
"""
self.validate_initialization()
self.prompt_template = ChatPromptTemplate.from_messages(
[
("system", system_template),
("human", human_template),
]
)
self.chain = self.prompt_template | self.llm.with_structured_output(output_model)
return self.chain
def build_return_with_tokens(self, node_specific_data: dict) -> dict:
"""
Comfort technique so as to add token utilization information into the return values.
"""
return {
**node_specific_data,
"prompt_tokens": self.prompt_tokens,
"completion_tokens": self.completion_tokens,
}
def __call__(self, state: DebateState) -> None:
"""
Updates the node's native copy of the state.
"""
self.state = state
for key, worth in state.gadgets():
setattr(self, key, worth)
Key options of BaseComponent:
- It shops an LLM shopper (e.g. an OpenAI
ChatOpenAIoccasion) initialized with a given mannequin and API key, in addition to an output parser. - It offers a way
create_chain(system_template, human_template)which units up a LangChain immediate chain (aRunnableSequence) combining a system immediate and a human immediate. This chain is what really generates outputs when run. - It has an
execute_chain(inputs)technique that invokes the chain and contains logic to retry if the OpenAI API returns a rate-limit error (HTTP 429). That is accomplished with exponential backoff as much as amax_retriesdepend. - It retains observe of token utilization (immediate tokens and completion tokens) for logging or evaluation.
- The
__call__technique of BaseComponent (which every subclass will name by way oftremendous().__call__(state)) can carry out any setup wanted earlier than the node’s fundamental logic runs (like guaranteeing the LLM is initialized).
By constructing on BaseComponent, every agent class can concentrate on its distinctive logic (like what immediate to make use of and learn how to deal with the state), whereas inheriting the heavy lifting of interacting with GPT-4o reliably.
Matter Generator Agent (GenerateTopicNode)
The Matter Generator (topic_generator_node.py) is the primary agent within the graph. Its job is to provide you with a debatable subject for the session. We give it a immediate that instructs it to output a nuanced subject that would moderately have a professional and con aspect.
This agent inherits from BaseComponent and makes use of a immediate chain (system + human immediate) to generate one merchandise of textual content – the controversy subject. When referred to as, it executes the chain (with no particular enter, simply utilizing the immediate) and will get again a topic_text. It then updates the state with:
debate_topic: the generated subject (stripped of any additional whitespace),positions: a dictionary assigning the professional and con stances (by default we use"In favor of the subject"and"Towards the subject"),stage: set to"opening",speaker: set to"professional"(so the Professional aspect will converse first).
In code, the return would possibly appear like:
return {
"debate_topic": debate_topic,
"positions": positions,
"stage": "opening",
"speaker": first_speaker # "professional"
}
Listed here are the prompts for the subject generator:
SYSTEM_PROMPT = """
You're a brainstorming AI that means debate subjects.
You'll present a single, attention-grabbing or well timed subject that may have two opposing views.
"""
HUMAN_PROMPT = """
Please recommend one debate subject for 2 AI brokers to debate.
For instance, it could possibly be about expertise, politics, philosophy, or any attention-grabbing area.
Simply present the subject in a concise sentence.
"""Then we go these prompts within the constructor of the category itself.
class GenerateTopicNode(BaseComponent):
def __init__(self, llm_config, temperature: float = 0.7):
tremendous().__init__(llm_config, temperature)
# Create the immediate chain.
self.chain: RunnableSequence = self.create_chain(
system_template=SYSTEM_PROMPT,
human_template=HUMAN_PROMPT
)
def __call__(self, state: DebateState) -> Dict[str, str]:
"""
Generates a debate subject and assigns positions to the 2 debaters.
"""
tremendous().__call__(state)
topic_text = self.execute_chain({})
# Retailer the subject and assign stances within the DebateState
debate_topic = topic_text.strip()
positions = {
"professional": "In favor of the subject",
"con": "Towards the subject"
}
first_speaker = "professional"
self.logger.information("Welcome to our debate panel! At this time's debate subject is: %s", debate_topic)
return {
"debate_topic": debate_topic,
"positions": positions,
"stage": "opening",
"speaker": first_speaker
}It’s a sample we are going to repeat for all lessons aside from these not utilizing LLMs and the actual fact checker.
Now we are able to implement the two stars of the present, the Professional and Con argument brokers!
Debater Brokers (Professional and Con)
Hyperlink: pro_debater_node.py
The 2 debater brokers are very comparable in construction, however every makes use of completely different immediate templates tailor-made to their function (professional vs con) and the stage of the controversy.
The Professional debater, for instance, has to deal with an opening assertion and a counter-argument (countering the Con’s rebuttal). We additionally want logic for retries in case an announcement fails fact-check. In code, the ProDebater class units up a number of immediate chains:
opening_chainand anopening_retry_chain(utilizing barely completely different human prompts – the retry immediate would possibly instruct it to strive once more with out repeating any factually doubtful claims).counter_chainandcounter_retry_chainfor the counter-argument stage.
class ProDebaterNode(BaseComponent):
def __init__(self, llm_config, temperature: float = 0.7):
tremendous().__init__(llm_config, temperature)
self.opening_chain = self.create_chain(SYSTEM_PROMPT, OPENING_HUMAN_PROMPT)
self.opening_retry_chain = self.create_chain(SYSTEM_PROMPT, OPENING_RETRY_HUMAN_PROMPT)
self.counter_chain = self.create_chain(SYSTEM_PROMPT, COUNTER_HUMAN_PROMPT)
self.counter_retry_chain = self.create_chain(SYSTEM_PROMPT, COUNTER_RETRY_HUMAN_PROMPT)
def __call__(self, state: DebateState) -> Dict[str, Any]:
tremendous().__call__(state)
debate_topic = state.get("debate_topic")
messages = state.get("messages", [])
stage = state.get("stage")
speaker = state.get("speaker")
# Verify if retrying (final message was by professional and never validated)
last_msg = messages[-1] if messages else None
retrying = last_msg and last_msg["speaker"] == SPEAKER_PRO and never last_msg["validated"]
if stage == STAGE_OPENING and speaker == SPEAKER_PRO:
chain = self.opening_retry_chain if retrying else self.opening_chain # choose which chain we're triggering: the conventional one or the fact-cehcked one
consequence = chain.invoke({
"debate_topic": debate_topic
})
elif stage == STAGE_COUNTER and speaker == SPEAKER_PRO:
opponent_msg = self._get_last_message_by(SPEAKER_CON, messages)
debate_history = get_debate_history(messages)
chain = self.counter_retry_chain if retrying else self.counter_chain
consequence = chain.invoke({
"debate_topic": debate_topic,
"opponent_statement": opponent_msg,
"debate_history": debate_history
})
else:
elevate ValueError(f"Unknown flip for ProDebater: stage={stage}, speaker={speaker}")
new_message = create_debate_message(speaker=SPEAKER_PRO, content material=consequence, stage=stage)
self.logger.information("Speaker: %s, Stage: %s, Retry: %snMessage:npercents", speaker, stage, retrying, consequence)
return {
"messages": messages + [new_message]
}
def _get_last_message_by(self, speaker_prefix, messages):
for m in reversed(messages):
if m.get("speaker") == speaker_prefix:
return m["content"]
return ""When the ProDebater’s __call__ runs, it seems on the present stage and speaker within the state to determine what to do:
- If it’s the opening stage and the speaker is “professional”, it makes use of the
opening_chainto generate a gap argument. If the final message from Professional was marked invalid (not validated), it is aware of it is a retry, so it will use theopening_retry_chainas an alternative. - If it’s the counter stage and speaker is “professional”, it generates a counter-argument to regardless of the opponent (Con) simply mentioned. It would fetch the final message by the Con from the
messageshistorical past, and feed that into the immediate (in order that the Professional can immediately counter it). Once more, if the final Professional message was invalid, it will change to the retry chain.
After producing its argument, the Debater agent creates a brand new message entry (with speaker="professional", the content material textual content, validated=False initially, and the stage) and appends it to the state’s message record. That turns into the output of the node (LangGraph will merge this partial state replace into the worldwide state).
The Con Debater agent mirrors this logic for its levels:
It equally appends its message to the state.
It has a rebuttal and closing argument (last argument) stage, every with a traditional and a retry chain.
It checks if it’s the rebuttal stage (speaker “con”) or last argument stage (speaker “con”) and invokes the suitable chain, presumably utilizing the final Professional message for context when rebutting.
Through the use of class-based implementation, our debaters’ code is simpler to keep up. We are able to clearly separate what the Professional does vs what the Con does, even when they share construction. Additionally, by encapsulating immediate chains inside the category, every debater can handle a number of doable outputs (common vs retry) cleanly.
Immediate design: The precise prompts (in prompts/pro_debater_prompts.py and con_debater_prompts.py) information the GPT-4o mannequin to tackle a persona (“You’re a debater arguing for/towards the subject…”) and produce the argument. In addition they instruct the mannequin to maintain statements factual and logical. If a truth verify fails, the retry immediate could say one thing like: “Your earlier assertion had an unverified declare. Revise your argument to be factually appropriate whereas sustaining your place.” – encouraging the mannequin to appropriate itself.
With this, our AI debaters can interact in a multi-turn duel, and even recuperate from factual missteps.
Truth Checker Agent (FactCheckNode)
After every debater speaks, the Truth Checker agent swoops in to confirm their claims. This agent is applied in <a href="https://github.com/iason-solomos/Deb8flow/blob/fundamental/nodes/fact_checker_node.py">fact_checker_node.py</a>, and apparently, it makes use of the GPT-4o mannequin’s shopping capacity somewhat than our personal customized prompts. Basically, we delegate the fact-checking to OpenAI’s GPT-4 with net search.
How does this work? The OpenAI Python shopper for GPT-4 (with shopping) permits us to ship a consumer message and get a structured response. In FactCheckNode.__call__, we do one thing like:
completion = self.shopper.beta.chat.completions.parse(
mannequin="gpt-4o-search-preview",
web_search_options={},
messages=[{
"role": "user",
"content": (
f"Consider the following statement from a debate. "
f"If the statement contains numbers, or figures from studies, fact-check it online.nn"
f"Statement:n"{claim}"nn"
f"Reply clearly whether any numbers or studies might be inaccurate or hallucinated, and why."
f"n"
f"If the statement doesn't contain references to studies or numbers cited, don't go online to fact-check, and just consider it successfully fact-checked, with a 'yes' score.nn"
)
}],
response_format=FactCheck
)If the result’s “sure” (which means the declare appears truthful or at the very least not factually mistaken), the Truth Checker will mark the final message’s validated area as True within the state, and output {"validated": True} with no additional adjustments. This indicators that the controversy can proceed usually.
If the result’s “no” (which means it discovered the declare to be incorrect or doubtful), the Truth Checker will append a brand new message to the state with speaker="fact_checker" describing the discovering (or we might merely mark it, however offering a quick notice like “(Truth Checker: The statistic cited couldn’t be verified.)” could be helpful). It would additionally set validated: False and increment a counter for whichever aspect made the declare. The output state from this node contains validated: False and an up to date times_pro_fact_checked or times_con_fact_checked depend.
We additionally use a Pydantic BaseModel to manage the output of the LLM:
class FactCheck(BaseModel):
"""
Pydantic mannequin for the actual fact checking the claims made by debaters.
Attributes:
binary_score (str): 'sure' if the declare is verifiable and truthful, 'no' in any other case.
"""
binary_score: str = Area(
description="Signifies if the declare is verifiable and truthful. 'sure' or 'no'."
)
justification: str = Area(
description="Rationalization of the reasoning behind the rating."
)Debate Moderator Agent (DebateModeratorNode)
The Debate Moderator is the conductor of the controversy. As an alternative of manufacturing prolonged textual content, this agent’s job is to handle turn-taking and stage development. Within the workflow, after an announcement is validated by the Truth Checker, management passes to the Moderator node. The Moderator then points a Command that updates the state for the following flip and directs the move to the suitable subsequent agent.
The logic in DebateModeratorNode.__call__ (see <a href="https://github.com/iason-solomos/Deb8flow/blob/fundamental/nodes/debate_moderator_node.py">nodes/debate_moderator_node.py</a>) goes roughly like this:
if stage == STAGE_OPENING and speaker == SPEAKER_PRO:
return Command(
replace={"stage": STAGE_REBUTTAL, "speaker": SPEAKER_CON},
goto=NODE_CON_DEBATER
)
elif stage == STAGE_REBUTTAL and speaker == SPEAKER_CON:
return Command(
replace={"stage": STAGE_COUNTER, "speaker": SPEAKER_PRO},
goto=NODE_PRO_DEBATER
)
elif stage == STAGE_COUNTER and speaker == SPEAKER_PRO:
return Command(
replace={"stage": STAGE_FINAL_ARGUMENT, "speaker": SPEAKER_CON},
goto=NODE_CON_DEBATER
)
elif stage == STAGE_FINAL_ARGUMENT and speaker == SPEAKER_CON:
return Command(
replace={},
goto=NODE_JUDGE
)
elevate ValueError(f"Surprising stage/speaker combo: stage={stage}, speaker={speaker}")Every conditional corresponds to a degree within the debate the place a flip simply ended, and units up the following flip. For instance, after the opening (Professional simply spoke), it units stage to rebuttal, switches speaker to Con, and directs the workflow to the Con debater node. After the final_argument (Con’s closing), it directs to the Choose with no additional replace (the controversy stage successfully ends).
Truth Verify Router (FactCheckRouterNode)
That is one other management node (just like the Moderator) that introduces conditional logic. The Truth Verify Router sits proper after the Truth Checker agent within the move. Its objective is to department the workflow relying on the fact-check consequence.
In <a href="https://github.com/iason-solomos/Deb8flow/blob/fundamental/nodes/fact_check_router_node.py">nodes/fact_check_router_node.py</a>, the logic is:
if pro_fact_checks >= 3 or con_fact_checks >= 3:
disqualified = SPEAKER_PRO if pro_fact_checks >= 3 else SPEAKER_CON
winner = SPEAKER_CON if disqualified == SPEAKER_PRO else SPEAKER_PRO
verdict_msg = {
"speaker": "moderator",
"content material": (
f"Debate ended early on account of extreme factual inaccuracies.nn"
f"DISQUALIFIED: {disqualified.higher()} (exceeded truth verify restrict)n"
f"WINNER: {winner.higher()}"
),
"validated": True,
"stage": "verdict"
}
return Command(
replace={"messages": messages + [verdict_msg]},
goto=END
)
if last_message.get("validated"):
return Command(goto=NODE_DEBATE_MODERATOR)
elif speaker == SPEAKER_PRO:
return Command(goto=NODE_PRO_DEBATER)
elif speaker == SPEAKER_CON:
return Command(goto=NODE_CON_DEBATER)
elevate ValueError("Unable to find out routing in FactCheckRouterNode.")First, the Truth Verify Router checks if both aspect’s fact-check depend has reached 3. If that’s the case, it creates a Moderator-style message saying an early finish: the offending aspect is disqualified and the opposite aspect is the winner. It appends this verdict to the messages and returns a Command that jumps to END, successfully terminating the controversy with out going to the Choose (as a result of we already know the result).
If we’re not ending the controversy early, it then seems on the Truth Checker’s consequence for the final message (which is saved as validated on that message). If validated is True, we go to the controversy moderator: Command(goto=debate_moderator_node).
Else if the assertion fails fact-check, the workflow goes again to the debater to provide a revised assertion (with the state counters up to date to mirror the failure). This loop can occur a number of occasions if wanted (as much as the disqualification restrict).
This dynamic management is the guts of Deb8flow’s “agentic” nature – the flexibility to adapt the trail of execution primarily based on the content material of the brokers’ outputs. It showcases LangGraph’s energy: combining management move with state. We’re basically encoding debate guidelines (like permitting retries for false claims, or ending the controversy if somebody cheats too usually) immediately into the workflow graph.
Choose Agent (JudgeNode)
Final however not least, the Choose agent delivers the ultimate verdict primarily based on rhetorical ability, readability, construction, and general persuasiveness. Its system immediate and human immediate make this express:
- System Immediate: “You might be an neutral debate choose AI. … Consider which debater introduced their case extra clearly, persuasively, and logically. You have to concentrate on communication expertise, construction of argument, rhetorical energy, and general coherence.”
- Human Immediate: “Right here is the complete debate transcript. Please analyze the efficiency of each debaters—PRO and CON. Consider rhetorical efficiency—readability, construction, persuasion, and relevance—and determine who introduced their case extra successfully.”
When the Choose node runs, it receives the complete debate transcript (all validated messages) alongside the unique subject. It then makes use of GPT-4o to look at how all sides framed their arguments, dealt with counterpoints, and supported (or did not help) claims with examples or logic. Crucially, the Choose is forbidden to judge which place is objectively appropriate (or who it thinks is likely to be appropriate)—solely who argued extra persuasively.
Beneath is an instance last verdict from a Deb8flow run on the subject:
“Ought to governments implement a common fundamental revenue in response to growing automation within the workforce?”
WINNER: PRO
REASON: The PRO debater introduced a extra compelling and rhetorically efficient case for common fundamental revenue. Their arguments have been well-structured, starting with a transparent assertion of the difficulty and the need of UBI in response to automation. They successfully addressed potential counterarguments by highlighting the unprecedented velocity and scope of present technological adjustments, which distinguishes the present scenario from previous technological shifts. The PRO additionally supplied empirical proof from UBI pilot packages to counter the CON's claims about work disincentives and financial inefficiencies, reinforcing their argument with real-world examples.
In distinction, the CON debater, whereas presenting legitimate issues about UBI, relied closely on historic analogies and assumptions about workforce adaptability with out adequately addressing the distinctive challenges posed by trendy automation. Their arguments in regards to the fiscal burden and potential inefficiencies of UBI have been much less supported by particular proof in comparison with the PRO's rebuttals.
Total, the PRO's arguments have been extra coherent, persuasive, and backed by empirical proof, making their case extra convincing to a impartial observer.
Langsmith Tracing
All through Deb8flow’s improvement, I relied on LangSmith (LangChain’s tracing and observability toolkit) to make sure the complete debate pipeline was behaving accurately. As a result of we have now a number of brokers passing management between themselves, it’s straightforward for sudden loops or misrouted states to happen. LangSmith offers a handy option to:
- Visualize Execution Movement: You may see every agent’s immediate, the tokens consumed (so you too can observe prices), and any intermediate states. This makes it a lot less complicated to substantiate that, say, the Con Debater is correctly referencing the Professional Debater’s final message, or that the Truth Checker is precisely receiving the declare to confirm.
- Debug State Updates: If the Moderator or Truth Verify Router is sending the move to the mistaken node, the hint will spotlight that mismatch. You may hint which agent was invoked at every step and why, serving to you see stage or speaker misalignments early.
- Observe Immediate and Completion Tokens: With a number of GPT-4o calls, it’s helpful to see what number of tokens every stage is utilizing, which LangSmith logs routinely should you allow tracing.
Integrating LangSmith is unexpectedly straightforward. You’ll simply want to offer these 3 keys in your .env file: LANGCHAIN_API_KEY
LANGCHAIN_TRACING_V2
LANGCHAIN_PROJECT
Then you possibly can open the LangSmith UI to see a structured hint of every run. This tremendously reduces the guesswork concerned in debugging multi-agent techniques and is, in my expertise, important for extra advanced AI orchestration like ours. Instance of a single run:
Reflections and Subsequent Steps
Constructing Deb8flow was an eye-opening train in orchestrating autonomous agent workflows. We didn’t simply chain a single mannequin name – we created a complete debate simulation with AI brokers, every with a particular function, and allowed them to work together in keeping with a algorithm. LangGraph supplied a transparent framework to outline how knowledge and management flows between brokers, making the advanced sequence manageable in code. Through the use of class-based brokers and a shared state, we maintained modularity and readability, which can repay for any software program engineering mission in the long term.
An thrilling side of this mission was seeing emergent habits. Though every agent follows a script (a immediate), the unscripted mixture – a debater attempting to deceive, a fact-checker catching it, the debater rephrasing – felt surprisingly life like! It’s a small step towards extra Agentic Ai techniques that may carry out non-trivial multi-step duties with oversight on one another.
There’s loads of concepts for enchancment:
- Consumer Interplay: At the moment it’s totally autonomous, however one might add a mode the place a human offers the subject and even takes the function of 1 aspect towards an AI opponent.
- We are able to change the order wherein the Debaters discuss.
- We are able to change the prompts, and thus to a very good diploma the habits of the brokers, and experiment with completely different prompts.
- Make the debaters additionally carry out net search earlier than producing their statements, thus offering them with the newest data.
The broader implication of Deb8flow is the way it showcases a sample for composable AI brokers. By defining clear boundaries and interactions (identical to microservices in software program), we are able to have advanced AI-driven processes that stay interpretable and controllable. Every agent is sort of a cog in a machine, and LangGraph is the gear system making them work in unison.
I discovered this mission energizing, and I hope it evokes you to discover multi-agent workflows. Whether or not it’s debating, collaborating on writing, or fixing issues from completely different knowledgeable angles, the mixture of GPT, instruments, and structured agentic workflows opens up a brand new world of potentialities for AI improvement. Joyful hacking!
References
[1] D. Bouchard, “From Fundamentals to Superior: Exploring LangGraph,” Medium, Nov. 22, 2023. [Online]. Out there: https://medium.com/data-science/from-basics-to-advanced-exploring-langgraph-e8c1cf4db787. [Accessed: Apr. 1, 2025].
[2] A. W. T. Ng, “Constructing a Analysis Agent that Can Write to Google Docs: Half 1,” In the direction of Information Science, Jan. 11, 2024. [Online]. Out there: https://towardsdatascience.com/building-a-research-agent-that-can-write-to-google-docs-part-1-4b49ea05a292/. [Accessed: Apr. 1, 2025].