langchain-ai/langchain · error · ValueError

Function must have a docstring if description not provided.

Error message

Function must have a docstring if description not provided.

What it means

A tool must carry a description for the model; from_function derives it from the function's docstring when description is not given. If the function has no docstring, no description kwarg, and no description in a dict args_schema, construction fails with ValueError.

Source

Thrown at libs/core/langchain_core/tools/structured.py:235

                description_ = args_schema.__doc__
                if (
                    description_
                    and "A base class for creating Pydantic models" in description_
                ):
                    description_ = ""
                elif not description_:
                    description_ = None
            elif isinstance(args_schema, dict):
                description_ = args_schema.get("description")
            else:
                msg = (
                    "Invalid args_schema: expected BaseModel or dict, "
                    f"got {args_schema}"
                )
                raise TypeError(msg)
        if description_ is None:
            msg = "Function must have a docstring if description not provided."
            raise ValueError(msg)
        if description is None:
            # Only apply if using the function's docstring
            description_ = textwrap.dedent(description_).strip()

        # Description example:
        # search_api(query: str) - Searches the API for the query.
        description_ = f"{description_.strip()}"
        return cls(
            name=name,
            func=func,
            coroutine=coroutine,
            args_schema=args_schema,
            description=description_,
            return_direct=return_direct,
            response_format=response_format,
            **kwargs,
        )

View on GitHub (pinned to e32fa9a52e)

Solutions

1. Pass description= explicitly to from_function
2. Or add a docstring to the source function/coroutine
3. When using a dict args_schema, include a 'description' key

Example fix

# before
tool = StructuredTool.from_function(
    lambda q: search(q), name="search"
)

# after
tool = StructuredTool.from_function(
    lambda q: search(q),
    name="search",
    description="Search the web for a query.",
)

import inspect

def described_from_function(fn, **kw):
    has_doc = bool(inspect.getdoc(fn))
    schema_desc = isinstance(kw.get("args_schema"), dict) and kw["args_schema"].get("description")
    if not (kw.get("description") or has_doc or schema_desc):
        msg = f"Tool {getattr(fn, '__name__', fn)!r} has no description or docstring"
        raise ValueError(msg)
    return StructuredTool.from_function(fn, **kw)

try:
    t = StructuredTool.from_function(fn, name=name)
except ValueError as e:
    if "docstring" in str(e):
        t = StructuredTool.from_function(fn, name=name, description=f"TODO: {name}")
    else:
        raise

When it happens

Trigger: StructuredTool.from_function(lambda x: x) (lambdas have no docstring) without description=; functions whose docstrings were stripped; dict args_schema lacking a 'description' key.

Common situations: Quick lambda-based tools; lint rules (e.g. D103) that don't require docstrings so devs omit them; automated tool generation from OpenAPI specs where the summary field was not mapped to description.

Related errors

• Function must have either a docstring or description when in
• Function and/or coroutine must be provided
• StructuredTool does not support sync invocation.
• Function and/or coroutine must be provided
• Runnable must have an object schema.

AI-assisted analysis of langchain-ai/langchain@e32fa9a52e (2026-08-14). Data as JSON: /api/errors/db3e21cdfa87a253. Report an issue: GitHub.