langchain-ai/langchain · error · ValueError

Path '{path}' is absolute. Absolute paths are not allowed wh

Error message

Path '{path}' is absolute. Absolute paths are not allowed when loading prompt configurations to prevent path traversal attacks. Use relative paths instead, or pass `allow_dangerous_paths=True` if you trust the input.

What it means

`_validate_path` in `langchain_core.prompts.loading` rejects absolute paths when loading prompt configuration files (default behavior, `allow_dangerous_paths=False`). This is a path-traversal defense: absolute paths let a malicious prompt config point the loader at arbitrary files on the host.

Source

Thrown at libs/core/langchain_core/prompts/loading.py:38

def _validate_path(path: Path) -> None:
    """Reject absolute paths and `..` traversal components.

    Args:
        path: The path to validate.

    Raises:
        ValueError: If the path is absolute or contains `..` components.
    """
    if path.is_absolute():
        msg = (
            f"Path '{path}' is absolute. Absolute paths are not allowed "
            f"when loading prompt configurations to prevent path traversal "
            f"attacks. Use relative paths instead, or pass "
            f"`allow_dangerous_paths=True` if you trust the input."
        )
        raise ValueError(msg)
    if ".." in path.parts:
        msg = (
            f"Path '{path}' contains '..' components. Directory traversal "
            f"sequences are not allowed when loading prompt configurations. "
            f"Use direct relative paths instead, or pass "
            f"`allow_dangerous_paths=True` if you trust the input."
        )
        raise ValueError(msg)


@deprecated(
    since="1.2.21",
    removal="2.0.0",
    alternative="Use `dumpd`/`dumps` from `langchain_core.load` to serialize "
    "prompts and `load`/`loads` to deserialize them.",
)
def load_prompt_from_config(
    config: dict[str, Any], *, allow_dangerous_paths: bool = False

View on GitHub (pinned to e32fa9a52e)

Solutions

1. Pass a path relative to your current working directory: `load_prompt('prompts/config.json')`.
2. If you fully trust the config source, opt in explicitly: `load_prompt(p, allow_dangerous_paths=True)`.
3. Change the config file so `template_path`/`example_prompt_path`/`examples` entries are relative to the config file's directory.

Example fix

# before
load_prompt('/abs/path/prompt.json')  # ValueError

# after (trusted input only)
load_prompt('/abs/path/prompt.json', allow_dangerous_paths=True)
# or use a relative path
load_prompt('prompts/prompt.json')

from pathlib import Path

p = Path(user_path)
if p.is_absolute():
    p = Path.cwd() / p  # or reject, depending on trust
load_prompt(str(p))

from pathlib import Path

def is_relative_safe(path: str | Path) -> bool:
    p = Path(path)
    return not p.is_absolute() and '..' not in p.parts

When it happens

Trigger: Calling `load_prompt('/home/user/prompts/config.json')` or a config with `template_path: /etc/passwd`-style absolute references without passing `allow_dangerous_paths=True`.

Common situations: Scripts that store prompt YAMLs at absolute locations (common in notebooks/containers); loading third-party or hub-downloaded prompt configs that reference templates by absolute path; CI pipelines with fixed workspace paths.

Related errors

• Path '{path}' contains '..' components. Directory traversal
• Loading templates with '{template_format}' format is no long
• jinja2 is unsafe and is not supported for templates expresse
• Loading images from 'path' has been removed as of 0.3.15 for
• Loading {config_type} prompt not supported

AI-assisted analysis of langchain-ai/langchain@e32fa9a52e (2026-08-14). Data as JSON: /api/errors/43ec007da5619692. Report an issue: GitHub.