ai-agents-tool-input-parsing-agent-fails-to-execute-tool-with-error-the-action--0bd2d55a

THEN Define tool's args_schema using a Pydantic BaseModel with explicit Field descriptions for each parameter. In the agent's task description, instruct the LLM to output a valid JSON object matching that schema without extra string escaping. Avoid passing JSON strings directly in tool config; let the LLM generate the raw dictionary.

class ArxivSearchInput(BaseModel):
    author: str = Field(description="Author name")
    title: str = Field(description="Paper title")
    max_results: int = Field(default=10)

class ArxivTool(BaseTool):
    name: str = "arxiv_search"
    args_schema: Type[BaseModel] = ArxivSearchInput
    def _run(self, **kwargs) -> str:
        return search(kwargs)

# In task description:
# 'Provide a valid JSON object with keys: author, title, max_results. Do not wrap in quotes.'