KeyError: 'arguments'

[njtraveller]

现在，我应该制定一个详细的计划，每个步骤都分配给合适的代理，并确保任务按时完成。

{
  "thought": "The user wants a detailed plan for a 30-day self-driving trip from Nanjing to Tibet, including accommodation arrangements and arrangement, cost estimation with a detailed breakdown.",
  "title": "Detailed Plan for 30-Day Self-Driving Trip from Nanjing to Tibet",
  "steps": [
    {
      "agent_name": "researcher",
      "title": "Research Travel Routes",
      "description": "Use search engines to gather information on major travel routes from Nanjing to Tibet, including key cities along the way. Focus on highways and scenic routes suitable for a self-driving trip.",
      "note": "Ensure information is up-to-date and includes any recent road closures or construction."
    },
    {
      "agent_name": "researcher",
      "title": "Accommodation Information Collection",
      "description": "Research and collect detailed information on hotels, hostels, and other accommodation options along the route. Include pricing, amenities, and reviews.",
      "note": "Focus on cities where the trip will spend more than one night."
    },
    {
      "agent_name": "researcher",
      "title": "Cost Estimation of Travel Expenses",
      "description": "Gather data on fuel costs, toll fees, vehicle maintenance expenses, and other travel-related expenses. Consider the current prices and usage rates.",
      "note": "Include average costs per kilometer for fuel and possible breakdowns."
    },
    {
      "agent_name": "coder",
      "title": "Calculate Total Estimated Costs",
      "description": "Use Python to calculate total estimated costs based on gathered data, including accommodation, food, transportation, and other expenses. Create a detailed cost breakdown in a table format.",
      "note": "Ensure all calculations are accurate and include a margin for unexpected expenses."
    },
    {
      "agent_name": "reporter",
      "title": "Prepare Final Report",
      "description": "Write a professional report summarizing the travel plan, accommodation arrangements, cost breakdown, and additional tips. Structure the report in an easy-to-read format.",
      "note": "Submit the report as the final output."
    }
  ]
}

INFO [LiteLLM] selected model name for cost calculation: ollama/deepseek-r1:14b
INFO [src.graph.nodes] Supervisor evaluating next action
20:02:28 - LiteLLM:INFO: cost_calculator.py:588 - selected model name for cost calculation: ollama/deepseek-r1:14b
20:02:28 - LiteLLM:INFO: utils.py:3035 -
LiteLLM completion() model= deepseek-r1:14b; provider = ollama
INFO [LiteLLM]
LiteLLM completion() model= deepseek-r1:14b; provider = ollama
function_call: {'next': 'researcher'}

Give Feedback / Get Help: https://github.com/BerriAI/litellm/issues/new
LiteLLM.Info: If you need to debug this error, use `litellm._turn_on_debug()'.

WARNING [langchain_core.language_models.llms] Retrying langchain_community.chat_models.litellm.ChatLiteLLM.completion_with_retry.._completion_with_retry in 4.0 seconds as it raised APIConnectionError: litellm.APIConnectionError: 'arguments'
Traceback (most recent call last):
File "C:\Users\Administrator.conda\envs\langmanus\Lib\site-packages\litellm\main.py", line 2832, in completion
response = base_llm_http_handler.completion(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\Administrator.conda\envs\langmanus\Lib\site-packages\litellm\llms\custom_httpx\llm_http_handler.py", line 388, in completion
return provider_config.transform_response(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\Administrator.conda\envs\langmanus\Lib\site-packages\litellm\llms\ollama\completion\transformation.py", line 273, in transform_response
"arguments": json.dumps(function_call["arguments"]),
~~~~~~~~~~~~~^^^^^^^^^^^^^
KeyError: 'arguments'
.