crewAIInc/crewAI · error · ValueError

Directory does not exist: {source_ref}

Error message

Directory does not exist: {source_ref}

What it means

ValueError raised when the resolved data type is DataType.DIRECTORY but os.path.isdir is False for the source reference: the adapter was told (or auto-detected) the input is a directory, and that directory does not exist at the given path.

Source

Thrown at lib/crewai-tools/src/crewai_tools/adapters/crewai_rag_adapter.py:213

            if isinstance(arg, dict):
                source_ref = str(arg.get("source", arg.get("content", "")))
            else:
                source_ref = str(arg)

            if not data_type:
                ext = os.path.splitext(source_ref)[1].lower()
                is_url = source_ref.startswith(("http://", "https://", "file://"))
                if (
                    ext in file_extensions
                    and not is_url
                    and not os.path.isfile(source_ref)
                ):
                    raise FileNotFoundError(f"File does not exist: {source_ref}")
                data_type = DataTypes.from_content(source_ref)

            if data_type == DataType.DIRECTORY:
                if not os.path.isdir(source_ref):
                    raise ValueError(f"Directory does not exist: {source_ref}")

                # Define binary and non-text file extensions to skip
                binary_extensions = {
                    ".pyc",
                    ".pyo",
                    ".png",
                    ".jpg",
                    ".jpeg",
                    ".gif",
                    ".bmp",
                    ".ico",
                    ".svg",
                    ".webp",
                    ".pdf",
                    ".zip",
                    ".tar",
                    ".gz",
                    ".bz2",

View on GitHub (pinned to 754d7323be)

Solutions

1. Use the absolute, resolved directory path and confirm it exists before ingestion.
2. Re-check configuration after moving a knowledge folder.
3. Validate at startup: for each configured source, assert os.path.isdir(source) or os.path.isfile(source).

Example fix

# before
adapter("./knowledge/base", data_type="directory")  # ValueError if missing

# after
from pathlib import Path
d = Path("./knowledge/base").resolve()
assert d.is_dir(), f"missing dir: {d}"
adapter(str(d), data_type=DataType.DIRECTORY)

import os

if data_type == DataType.DIRECTORY and not os.path.isdir(source_ref):
    raise ValueError(f"configured knowledge dir missing: {source_ref}")
adapter(source_ref, data_type=data_type)

def directory_source_ok(source_ref: str) -> bool:
    import os
    return os.path.isdir(source_ref)

try:
    adapter(source_ref, data_type=DataType.DIRECTORY)
except ValueError as e:
    if "Directory does not exist" in str(e):
        logger.error("fix configured knowledge dir path")
    raise

When it happens

Trigger: Passing a directory path whose data_type resolves to directory (e.g. via auto-detection of a path with no file extension) while the directory is missing, renamed, or the path is relative to a different cwd.

Common situations: Knowledge-source directory moved or renamed after configuration; relative directory paths in services with unpredictable cwd; trailing characters/typos in configured paths.

Related errors

• Invalid data_type: '{raw_data_type}'. Valid values are: 'fil
• File does not exist: {source_ref}
• URL directory loading is not supported. Please provide a loc
• Invalid --definition path: {definition} is not a file.
• Invalid --definition path: {definition} does not exist.

AI-assisted analysis of crewAIInc/crewAI@754d7323be (2026-08-15). Data as JSON: /api/errors/e652563a12d285fd. Report an issue: GitHub.