有run起来的吗，一直在plan循环 #10
Comments
|
建议planning agent使用claude3.7，目前发现gemini和gpt4o等模型是严格遵循function calling，不会使用json调用子agent，针对这个问题我后面会把子agent也通过function calling调用 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Plan: Find and summarize the latest AI Agent research paper (ID: ai_agent_research_plan)
Progress: 0/3 steps completed (0.0%)
Status: 0 completed, 1 in progress, 0 blocked, 2 not started
Steps:
0. [→] Use deep_researcher_agent to find the latest research paper on 'AI Agent'.
2025-05-27 17:18:02 - logger:INFO: logger.py:77 - Observations: Step 0 updated successfully in plan 'ai_agent_research_plan'.
Plan: Find and summarize the latest AI Agent research paper (ID: ai_agent_research_plan)
Progress: 0/3 steps completed (0.0%)
Status: 0 completed, 1 in progress, 0 blocked, 2 not started
Steps:
0. |→] Use deep_researcher_agent to find the latest research paper on 'AI Agent'.
2025-05-27 17:18:02 - logger:INFO: logger.py:77 - [Step 15: Duration 4.23 seconds| Input tokens: 68,296 | Output tokens: 1,973]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ Step 16 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Output message of the LLM: ─────────────────────────────────────────────────────────────────────────
ModelResponse(id='chatcmpl-30b78c6d-c2f0-408b-90e2-095f637d3bf3', created=1748337482,
model='gemini-2.5-pro-preview-05-06', object='chat.completion', system_fingerprint=None,
choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=None,
role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=0,
function=Function(arguments='{"plan_id": "ai_agent_research_plan", "step_index": 0, "action":
"mark_step", "step_status": "in_progress"}', name='planning'),
id='call_10d2456b-19b9-43d5-9e93-922c05ec2186', type='function')], function_call=None,
provider_specific_fields=None))], usage=Usage(completion_tokens=241, prompt_tokens=6930,
total_tokens=7171,
completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None,
audio_tokens=None, reasoning_tokens=197, rejected_prediction_tokens=None, text_tokens=44),
prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None,
text_tokens=6930, image_tokens=None)), vertex_ai_grounding_metadata=[], vertex_ai_safety_results=[],
vertex_ai_citation_metadata=[])
╭──────────────────────────────────────────────────────────────────────────────────────────────────╮
│ Calling tool: 'planning' with arguments: {'plan_id': 'ai_agent_research_plan', 'step_index': 0, │
│ 'action': 'mark_step', 'step_status': 'in_progress'} │
╰──────────────────────────────────────────────────────────────────────────────────────────────────╯
2025-05-27 17:18:08 - logger:INFO: planning.py:361 - Step 0 updated successfully in plan 'ai_agent_research_plan'.
Plan: Find and summarize the latest AI Agent research paper (ID: ai_agent_research_plan)
Progress: 0/3 steps completed (0.0%)
Status: 0 completed, 1 in progress, 0 blocked, 2 not started
Steps:
0. [→] Use deep_researcher_agent to find the latest research paper on 'AI Agent'.
2025-05-27 17:18:08 - logger:INFO: logger.py:77 - Observations: Step 0 updated successfully in plan 'ai_agent_research_plan'.
Plan: Find and summarize the latest AI Agent research paper (ID: ai_agent_research_plan)
Progress: 0/3 steps completed (0.0%)
Status: 0 completed, 1 in progress, 0 blocked, 2 not started
Steps:
0. |→] Use deep_researcher_agent to find the latest research paper on 'AI Agent'.
2025-05-27 17:18:08 - logger:INFO: logger.py:77 - [Step 16: Duration 5.76 seconds| Input tokens: 75,226 | Output tokens: 2,214]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ Step 17 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Output message of the LLM: ─────────────────────────────────────────────────────────────────────────
ModelResponse(id='chatcmpl-e0e5f5d6-d749-4beb-8680-d4f0161671dc', created=1748337488,
model='gemini-2.5-pro-preview-05-06', object='chat.completion', system_fingerprint=None,
choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=None,
role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=0,
function=Function(arguments='{"action": "mark_step", "step_index": 0, "plan_id":
"ai_agent_research_plan", "step_status": "in_progress"}', name='planning'),
id='call_9b5d9aa4-9c30-49ed-977b-7497f052db38', type='function')], function_call=None,
provider_specific_fields=None))], usage=Usage(completion_tokens=143, prompt_tokens=7228,
total_tokens=7371,
completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None,
audio_tokens=None, reasoning_tokens=99, rejected_prediction_tokens=None, text_tokens=44),
prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None,
text_tokens=7228, image_tokens=None)), vertex_ai_grounding_metadata=[], vertex_ai_safety_results=[],
vertex_ai_citation_metadata=[])
╭──────────────────────────────────────────────────────────────────────────────────────────────────╮
│ Calling tool: 'planning' with arguments: {'action': 'mark_step', 'step_index': 0, 'plan_id': │
│ 'ai_agent_research_plan', 'step_status': 'in_progress'} │
╰──────────────────────────────────────────────────────────────────────────────────────────────────╯
2025-05-27 17:18:13 - logger:INFO: planning.py:361 - Step 0 updated successfully in plan 'ai_agent_research_plan'.
Plan: Find and summarize the latest AI Agent research paper (ID: ai_agent_research_plan)
Progress: 0/3 steps completed (0.0%)
Status: 0 completed, 1 in progress, 0 blocked, 2 not started
Steps:
0. [→] Use deep_researcher_agent to find the latest research paper on 'AI Agent'.
2025-05-27 17:18:13 - logger:INFO: logger.py:77 - Observations: Step 0 updated successfully in plan 'ai_agent_research_plan'.
Plan: Find and summarize the latest AI Agent research paper (ID: ai_agent_research_plan)
Progress: 0/3 steps completed (0.0%)
Status: 0 completed, 1 in progress, 0 blocked, 2 not started
Steps:
0. |→] Use deep_researcher_agent to find the latest research paper on 'AI Agent'.
2025-05-27 17:18:13 - logger:INFO: logger.py:77 - [Step 17: Duration 4.73 seconds| Input tokens: 82,454 | Output tokens: 2,357]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ Step 18 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Output message of the LLM: ─────────────────────────────────────────────────────────────────────────
ModelResponse(id='chatcmpl-3ff43a23-7cfb-4ba3-b8f2-56108bf5d65a', created=1748337493,
model='gemini-2.5-pro-preview-05-06', object='chat.completion', system_fingerprint=None,
choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=None,
role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=0,
function=Function(arguments='{"step_index": 0, "plan_id": "ai_agent_research_plan", "step_status":
"in_progress", "action": "mark_step"}', name='planning'),
id='call_855068ee-b756-46d2-863b-08264d9b09ef', type='function')], function_call=None,
provider_specific_fields=None))], usage=Usage(completion_tokens=191, prompt_tokens=7524,
total_tokens=7715,
completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None,
audio_tokens=None, reasoning_tokens=147, rejected_prediction_tokens=None, text_tokens=44),
prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None,
text_tokens=7524, image_tokens=None)), vertex_ai_grounding_metadata=[], vertex_ai_safety_results=[],
vertex_ai_citation_metadata=[])
╭──────────────────────────────────────────────────────────────────────────────────────────────────╮
│ Calling tool: 'planning' with arguments: {'step_index': 0, 'plan_id': 'ai_agent_research_plan', │
│ 'step_status': 'in_progress', 'action': 'mark_step'} │
╰──────────────────────────────────────────────────────────────────────────────────────────────────╯
2025-05-27 17:18:18 - logger:INFO: planning.py:361 - Step 0 updated successfully in plan 'ai_agent_research_plan'.
Plan: Find and summarize the latest AI Agent research paper (ID: ai_agent_research_plan)
Progress: 0/3 steps completed (0.0%)
Status: 0 completed, 1 in progress, 0 blocked, 2 not started
Steps:
0. [→] Use deep_researcher_agent to find the latest research paper on 'AI Agent'.
2025-05-27 17:18:18 - logger:INFO: logger.py:77 - Observations: Step 0 updated successfully in plan 'ai_agent_research_plan'.
Plan: Find and summarize the latest AI Agent research paper (ID: ai_agent_research_plan)
Progress: 0/3 steps completed (0.0%)
Status: 0 completed, 1 in progress, 0 blocked, 2 not started
Steps:
0. |→] Use deep_researcher_agent to find the latest research paper on 'AI Agent'.
2025-05-27 17:18:18 - logger:INFO: logger.py:77 - [Step 18: Duration 5.10 seconds| Input tokens: 89,978 | Output tokens: 2,548]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ Step 19 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Output message of the LLM: ─────────────────────────────────────────────────────────────────────────
ModelResponse(id='chatcmpl-92bc9b05-b27c-4641-b3f7-fefc6bad4ec0', created=1748337498,
model='gemini-2.5-pro-preview-05-06', object='chat.completion', system_fingerprint=None,
choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=None,
role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=0,
function=Function(arguments='{"step_index": 0, "step_status": "in_progress", "plan_id":
"ai_agent_research_plan", "action": "mark_step"}', name='planning'),
id='call_ecdb8f81-adda-4ca3-83d0-d4ef33c70cf2', type='function')], function_call=None,
provider_specific_fields=None))], usage=Usage(completion_tokens=105, prompt_tokens=7821,
total_tokens=7926,
completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None,
audio_tokens=None, reasoning_tokens=61, rejected_prediction_tokens=None, text_tokens=44),
prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None,
text_tokens=7821, image_tokens=None)), vertex_ai_grounding_metadata=[], vertex_ai_safety_results=[],
vertex_ai_citation_metadata=[])
╭──────────────────────────────────────────────────────────────────────────────────────────────────╮
│ Calling tool: 'planning' with arguments: {'step_index': 0, 'step_status': 'in_progress', │
│ 'plan_id': 'ai_agent_research_plan', 'action': 'mark_step'} │
╰──────────────────────────────────────────────────────────────────────────────────────────────────╯
2025-05-27 17:18:22 - logger:INFO: planning.py:361 - Step 0 updated successfully in plan 'ai_agent_research_plan'.
Plan: Find and summarize the latest AI Agent research paper (ID: ai_agent_research_plan)
Progress: 0/3 steps completed (0.0%)
Status: 0 completed, 1 in progress, 0 blocked, 2 not started
Steps:
0. [→] Use deep_researcher_agent to find the latest research paper on 'AI Agent'.
2025-05-27 17:18:22 - logger:INFO: logger.py:77 - Observations: Step 0 updated successfully in plan 'ai_agent_research_plan'.
Plan: Find and summarize the latest AI Agent research paper (ID: ai_agent_research_plan)
Progress: 0/3 steps completed (0.0%)
Status: 0 completed, 1 in progress, 0 blocked, 2 not started
Steps:
0. |→] Use deep_researcher_agent to find the latest research paper on 'AI Agent'.
2025-05-27 17:18:22 - logger:INFO: logger.py:77 - [Step 19: Duration 4.36 seconds| Input tokens: 97,799 | Output tokens: 2,653]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ Step 20 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Output message of the LLM: ─────────────────────────────────────────────────────────────────────────
ModelResponse(id='chatcmpl-e5eeb4f6-fb86-43cd-aba4-e2ea104ec072', created=1748337502,
model='gemini-2.5-pro-preview-05-06', object='chat.completion', system_fingerprint=None,
choices=[Choices(finish_reason='tool_calls', index=0, message=Message(content=None,
role='assistant', tool_calls=[ChatCompletionMessageToolCall(index=0,
function=Function(arguments='{"step_status": "in_progress", "step_index": 0, "action": "mark_step",
"plan_id": "ai_agent_research_plan"}', name='planning'),
id='call_e92f19e9-754c-4eb6-b7d9-4304816655c3', type='function')], function_call=None,
provider_specific_fields=None))], usage=Usage(completion_tokens=138, prompt_tokens=8112,
total_tokens=8250,
completion_tokens_details=CompletionTokensDetailsWrapper(accepted_prediction_tokens=None,
audio_tokens=None, reasoning_tokens=94, rejected_prediction_tokens=None, text_tokens=44),
prompt_tokens_details=PromptTokensDetailsWrapper(audio_tokens=None, cached_tokens=None,
text_tokens=8112, image_tokens=None)), vertex_ai_grounding_metadata=[], vertex_ai_safety_results=[],
vertex_ai_citation_metadata=[])
╭──────────────────────────────────────────────────────────────────────────────────────────────────╮
│ Calling tool: 'planning' with arguments: {'step_status': 'in_progress', 'step_index': 0, │
│ 'action': 'mark_step', 'plan_id': 'ai_agent_research_plan'} │
╰──────────────────────────────────────────────────────────────────────────────────────────────────╯
2025-05-27 17:18:26 - logger:INFO: planning.py:361 - Step 0 updated successfully in plan 'ai_agent_research_plan'.
Plan: Find and summarize the latest AI Agent research paper (ID: ai_agent_research_plan)
Progress: 0/3 steps completed (0.0%)
Status: 0 completed, 1 in progress, 0 blocked, 2 not started
Steps:
0. [→] Use deep_researcher_agent to find the latest research paper on 'AI Agent'.
2025-05-27 17:18:26 - logger:INFO: logger.py:77 - Observations: Step 0 updated successfully in plan 'ai_agent_research_plan'.
Plan: Find and summarize the latest AI Agent research paper (ID: ai_agent_research_plan)
2025-05-27 17:18:26 - logger:INFO: logger.py:77 - Observations: Step 0 updated successfully in plan 'ai_agent_research_plan'.
Plan: Find and summarize the latest AI Agent research paper (ID: ai_agent_research_plan)
2025-05-27 17:18:26 - logger:INFO: logger.py:77 - Observations: Step 0 updated successfully in plan 'ai_agent_research_plan'.
Plan: Find and summarize the latest AI Agent research paper (ID: ai_agent_research_plan)
2025-05-27 17:18:26 - logger:INFO: logger.py:77 - Observations: Step 0 updated successfully in plan 'ai_agent_research_plan'.
2025-05-27 17:18:26 - logger:INFO: logger.py:77 - Observations: Step 0 updated successfully in plan 'ai_agent_research_plan'.
Plan: Find and summarize the latest AI Agent research paper (ID: ai_agent_research_plan)
2025-05-27 17:18:26 - logger:INFO: logger.py:77 - Observations: Step 0 updated successfully in plan 'ai_agent_research_plan'.
Plan: Find and summarize the latest AI Agent research paper (ID: ai_agent_research_plan)
2025-05-27 17:18:26 - logger:INFO: logger.py:77 - Observations: Step 0 updated successfully in plan 'ai_agent_research_plan'.
Plan: Find and summarize the latest AI Agent research paper (ID: ai_agent_research_plan)
2025-05-27 17:18:26 - logger:INFO: logger.py:77 - Observations: Step 0 updated successfully in plan 'ai_agent_research_plan'.
Plan: Find and summarize the latest AI Agent research paper (ID: ai_agent_research_plan)
2025-05-27 17:18:26 - logger:INFO: logger.py:77 - Observations: Step 0 updated successfully in plan 'ai_agent_research_plan'.
Plan: Find and summarize the latest AI Agent research paper (ID: ai_agent_research_plan)
Plan: Find and summarize the latest AI Agent research paper (ID: ai_agent_research_plan)
Progress: 0/3 steps completed (0.0%)
Status: 0 completed, 1 in progress, 0 blocked, 2 not started
Steps:
0. |→] Use deep_researcher_agent to find the latest research paper on 'AI Agent'.
2025-05-27 17:18:26 - logger:INFO: logger.py:77 - [Step 20: Duration 3.83 seconds| Input tokens: 105,911 | Output tokens: 2,791]
2025-05-27 17:18:26 - logger:ERROR: error.py:10 - Reached max steps.
2025-05-27 17:18:26 - logger:INFO: logger.py:77 - [Step 21: Duration 3.84 seconds| Input tokens: 114,023 | Output tokens: 2,929]
2025-05-27 17:18:26 - logger:INFO: test.py:31 - Result: <coroutine object AsyncMultiStepAgent.provide_final_answer at 0x000001CFC0ACB520>
The text was updated successfully, but these errors were encountered: