binary-husky/gpt_academic · error · Exception

用户代理或助理代理未定义

Error message

用户代理或助理代理未定义

What it means

AutoGenGeneral.exe_autogen() builds agents from define_agents() and recognizes the required agents by the exact names 'user_proxy' and 'assistant'. If either variable is still None after the loop, this Exception is raised, converted to a traceback, and sent back through the pipe.

Source

Thrown at crazy_functions/agent_fns/general.py:90

        assistant = None
        for agent_kwargs in agents:
            agent_cls = agent_kwargs.pop('cls')
            kwargs = {
                'llm_config':self.llm_kwargs,
                'code_execution_config':code_execution_config
            }
            kwargs.update(agent_kwargs)
            agent_handle = agent_cls(**kwargs)
            agent_handle._print_received_message = lambda a,b: self.gpt_academic_print_override(agent_kwargs, a, b)
            for d in agent_handle._reply_func_list:
                if hasattr(d['reply_func'],'__name__') and d['reply_func'].__name__ == 'generate_oai_reply':
                    d['reply_func'] = gpt_academic_generate_oai_reply
            if agent_kwargs['name'] == 'user_proxy':
                agent_handle.get_human_input = lambda a: self.gpt_academic_get_human_input(user_proxy, a)
                user_proxy = agent_handle
            if agent_kwargs['name'] == 'assistant': assistant = agent_handle
        try:
            if user_proxy is None or assistant is None: raise Exception("用户代理或助理代理未定义")
            with ProxyNetworkActivate("AutoGen"):
                user_proxy.initiate_chat(assistant, message=input)
        except Exception as e:
            tb_str = '```\n' + trimmed_format_exc() + '```'
            self.child_conn.send(PipeCom("done", "AutoGen 执行失败: \n\n" + tb_str))

    def subprocess_worker(self, child_conn):
        # ⭐⭐ run in subprocess
        self.child_conn = child_conn
        while True:
            msg = self.child_conn.recv()  # PipeCom
            self.exe_autogen(msg)


class AutoGenGroupChat(AutoGenGeneral):
    def exe_autogen(self, input):
        # ⭐⭐ run in subprocess
        import autogen

View on GitHub (pinned to d6bde0fa54)

Solutions

1. Make define_agents() return one agent dict with name='user_proxy' and one with name='assistant'.
2. Check the dict keys before initiate_chat and fail with the names that were found.
3. Use exact lowercase strings; do not include whitespace or labels.
4. Keep the agent class under 'cls' and do not consume the name key before this loop.
5. Write a unit test for define_agents()' returned names.

Example fix

# before
agents = self.define_agents()
for agent_kwargs in agents:
    ...
if user_proxy is None or assistant is None:
    raise Exception("用户代理或助理代理未定义")

# after
agents = self.define_agents()
names = {a.get("name") for a in agents}
missing = {"user_proxy", "assistant"} - names
if missing:
    raise ValueError(f"Missing required agents {missing}; found {names}")

required = {"user_proxy", "assistant"}
names = {agent.get("name") for agent in self.define_agents()}
missing = required - names
if missing:
    raise ValueError(f"define_agents() is missing required names: {sorted(missing)}")

def has_required_two_agent_names(agents) -> bool:
    names = {a.get("name") for a in agents if isinstance(a, dict)}
    return {"user_proxy", "assistant"}.issubset(names)

try:
    ...initiate_chat...
except Exception as e:
    if "用户代理或助理代理未定义" in str(e):
        fix_agent_configuration()
    else:
        raise

When it happens

Trigger: A subclass's define_agents() returns no list entry with name exactly 'user_proxy' or 'assistant', renames them to values such as 'user proxy', 'proxy', or 'coder', or returns an empty list.

Common situations: Writing a custom agent plugin without following the base-class contract; changing names for display purposes; copying a group-chat configuration into the two-agent runner; conditionally removing an agent.

Related errors

• user_proxy is not defined
• 模型覆盖参数 '{model_override}' 指向一个暂不支持的模型，请检查配置文件。
• Endpoint不正确, 请检查AZURE_ENDPOINT的配置! 当前的Endpoint为:{endpoint}
• APIKEY为空,请检查配置文件的{APIKEY}
• 不支持的检索类型

AI-assisted analysis of binary-husky/gpt_academic@d6bde0fa54 (2026-08-14). Data as JSON: /api/errors/7dd656a04cebe23e. Report an issue: GitHub.