langchain_core.exceptions.ToolException: 工具 'search_tool' 调用缺少必需参数。期望：['query']，实际：[]。
langchain_core.exceptions.ToolException: Tool 'search_tool' called with missing required arguments. Expected: ['query'], got: [].
ID: llm/langchain-tool-call-missing-arguments
版本兼容性
| 版本 | 状态 | 引入 | 弃用 | 备注 |
|---|---|---|---|---|
| langchain==0.3.0 | active | — | — | — |
| langchain-core==0.3.5 | active | — | — | — |
| langchain-openai==0.2.0 | active | — | — | — |
根因分析
LLM 生成一个参数字典为空或缺少必需参数的工具调用，通常是由于模型误解了工具架构，或在流式传输期间截断了函数调用 JSON。
English
The LLM generates a tool call with an empty arguments dict or missing required parameters, often due to the model misinterpreting the tool schema or truncating the function call JSON during streaming.
官方文档
https://python.langchain.com/docs/how_to/custom_tools/#tool-validation解决方案
-
实现一个验证包装器，捕获 ToolException 并使用更正消息重新提示 LLM：`try: result = tool.run(tool_call) except ToolException: corrected_call = llm.invoke("之前的工具调用无效。请为 'search_tool' 提供有效的参数，包含 'query' 字符串。")`。这允许模型自我纠正。 -
通过使用原始类型（str、int）而不是复杂对象或枚举来简化工具架构。示例：将 `query: Optional[str]` 改为 `query: str = Field(description="...")` 使其成为必需且明确的参数。
-
使用 LangChain 的 `StructuredTool` 并设置 `handle_tool_error=True` 以自动重试并回退消息：`tool = StructuredTool.from_function(func=search, handle_tool_error=True)`。
无效尝试
常见但无效的做法:
-
65% 失败
While helpful, verbose descriptions can confuse the LLM or lead to token limit issues. The root cause is often the model's inability to consistently produce valid JSON for complex schemas.
-
70% 失败
Deterministic settings reduce variability but don't fix the underlying schema parsing issue. The model may still generate malformed calls if the prompt or tool definitions are ambiguous.
-
90% 失败
Retrying the same request with the same configuration will likely produce the same error because the model's behavior is consistent for the same input.