What causes "The schema is required for CodeGeneratorGraph" in ScrapeGraphAI/Scrapegraph-ai?

Pass a Pydantic BaseModel class (or JSON schema dict, per the graph's docs) as the schema argument.

ScrapeGraphAI/Scrapegraph-ai · error · KeyError

The schema is required for CodeGeneratorGraph

Error message

The schema is required for CodeGeneratorGraph

What it means

CodeGeneratorGraph._create_graph requires a schema because its whole purpose is generating scraping code that extracts fields described by the schema; with self.schema None it cannot build the prompt and raises KeyError('The schema is required for CodeGeneratorGraph') before constructing nodes.

Source

Thrown at scrapegraphai/graphs/code_generator_graph.py:78

        prompt: str,
        source: str,
        config: dict,
        schema: Optional[Type[BaseModel]] = None,
    ):
        super().__init__(prompt, config, source, schema)

        self.input_key = "url" if source.startswith("http") else "local_dir"

    def _create_graph(self) -> BaseGraph:
        """
        Creates the graph of nodes representing the workflow for web scraping.

        Returns:
            BaseGraph: A graph instance representing the web scraping workflow.
        """

        if self.schema is None:
            raise KeyError("The schema is required for CodeGeneratorGraph")

        fetch_node = FetchNode(
            input="url| local_dir",
            output=["doc"],
            node_config={
                "llm_model": self.llm_model,
                "force": self.config.get("force", False),
                "cut": self.config.get("cut", True),
                "loader_kwargs": self.config.get("loader_kwargs", {}),
                "browser_base": self.config.get("browser_base"),
                "scrape_do": self.config.get("scrape_do"),
                "storage_state": self.config.get("storage_state"),
            },
        )
        parse_node = ParseNode(
            input="doc",
            output=["parsed_doc"],
            node_config={

View on GitHub (pinned to 532dfffbf6)

Solutions

Pass a Pydantic BaseModel class (or JSON schema dict, per the graph's docs) as the schema argument.
If you don't need structured extraction, use ScriptGeneratorGraph or SmartScraperGraph instead, which do not require a schema.

Example fix

# before
graph = CodeGeneratorGraph(prompt='Extract books', source=url, config=config)

# after
class Books(BaseModel):
    title: str
    price: str
graph = CodeGeneratorGraph(prompt='Extract books', source=url, schema=Books, config=config)

Defensive patterns

Strategy: validation

Validate before calling

if schema is None:
    raise SystemExit('CodeGeneratorGraph requires a Pydantic schema describing the fields to extract')

Type guard

from pydantic import BaseModel
from typing import Type

def schema_ok(schema) -> bool:
    return schema is not None and (isinstance(schema, dict) or (isinstance(schema, type) and issubclass(schema, BaseModel)))

Try / catch

try:
    graph = CodeGeneratorGraph(prompt=p, source=url, schema=schema, config=cfg)
except KeyError as e:
    if 'schema is required' in str(e):
        raise SystemExit('Provide schema=YourPydanticModel') from e
    raise

Prevention

Always pass a schema to CodeGeneratorGraph — it is the core input, not an option.
Define extraction schemas as Pydantic models shared across your codebase.
Use SmartScraperGraph when no schema is needed.

When it happens

Trigger: Instantiating CodeGeneratorGraph(prompt, source, config) without the schema argument (defaults to None), or passing schema=None explicitly.

Common situations: Copy-pasting a SmartScraperGraph example (where schema is optional) into CodeGeneratorGraph; forgetting the third positional argument; passing the schema inside config instead of as the schema parameter.

Related errors

AI-assisted analysis of ScrapeGraphAI/Scrapegraph-ai@532dfffbf6 (2026-08-28). Data as JSON: /api/errors/96713fdb2af255c2. Report an issue: GitHub.