langchain-ai/deepagents · error · FileNotFoundError

Could not find {target_name} inside {archive.name}

Error message

Could not find {target_name} inside {archive.name}

What it means

`_extract_rg` raises `FileNotFoundError` when, after extracting the archive, no file named `rg` (or `rg.exe` on Windows) is found anywhere under the extraction root. This catches archives that are not the expected ripgrep release layout.

Source

Thrown at libs/code/deepagents_code/managed_tools.py:631

    Raises:
        FileNotFoundError: When the archive does not contain an `rg` binary.
    """
    import tarfile
    import zipfile

    if archive.suffix == ".zip":
        with zipfile.ZipFile(archive) as zf:
            _extract_zip_validated(zf, extract_root)
    else:
        with tarfile.open(archive, mode="r:*") as tf:
            tf.extractall(extract_root, filter="data")

    target_name = "rg.exe" if sys.platform == "win32" else "rg"
    for path in extract_root.rglob(target_name):
        if path.is_file():
            return path
    msg = f"Could not find {target_name} inside {archive.name}"
    raise FileNotFoundError(msg)


def _extract_zip_validated(zf: zipfile.ZipFile, extract_root: Path) -> None:
    """Extract a zip archive after validating each member's path.

    `ZipFile.extractall` does sanitize absolute paths and parent-relative
    components on modern Python, but defense-in-depth here keeps the
    SHA-256-verified archive from being the only line of defense against
    a zip-slip variant in a future upstream archive.

    Raises:
        zipfile.BadZipFile: If a member would extract outside `extract_root`.
    """
    import zipfile

    extract_root.mkdir(parents=True, exist_ok=True)
    root = extract_root.resolve()
    for member in zf.infolist():

View on GitHub (pinned to a1af029e6e)

Solutions

1. Confirm the downloaded archive matches the current platform/arch and re-download the correct release asset
2. Inspect the archive (`tar -tzf <archive>`) to verify `rg` exists inside; if the layout changed, update the extraction/lookup logic in `_extract_rg`
3. If the file is actually an HTML/error page, fix the download URL and proxy configuration

Example fix

// before
$ tar -tzf rg.tar.gz | head
musl-tools/   # wrong asset: no rg binary
// after
$ tar -tzf rg-14.1.0-x86_64-unknown-linux-musl.tar.gz | grep /rg$
rg-14.1.0-x86_64-unknown-linux-musl/rg
$ dcode install-tools   # with correct asset URL

import tarfile
from pathlib import Path

def archive_contains(archive: Path, member_suffix: str) -> bool:
    with tarfile.open(archive) as tf:
        return any(m.name.endswith(member_suffix) for m in tf.getmembers())

if not archive_contains(Path("rg.tar.gz"), "/rg"):
    raise SystemExit("wrong release asset: no rg binary inside")

try:
    rg = _extract_rg(archive, extract_root)
except FileNotFoundError as exc:
    raise SystemExit(f"bad tool archive: {exc}; re-download correct platform asset")

When it happens

Trigger: `_install_ripgrep_sync` (or the direct test path) given an archive whose contents lack the binary — a wrong-architecture or wrong-platform release tarball, an HTML error page saved as `.tar.gz`, or an archive whose nested layout changed between ripgrep versions.

Common situations: Downloading the musl vs gnu vs macOS wrong variant; a proxy serving an error page with a `.tar.gz` name; upstream restructuring the archive's inner directories; on Windows, downloading the non-`windows` asset so only `rg` (no `rg.exe`) is present.

Related errors

• Plugin source directory not found: {source}
• debug log directory is not a real directory: {path}
• debug log directory is not owned by the current user: {path}
• Cannot determine whether {str(left)!r} is {str(right)!r}: {e
• Invalid DEEPAGENTS_HOME {str(root)!r}: exists but cannot be

AI-assisted analysis of langchain-ai/deepagents@a1af029e6e (2026-08-29). Data as JSON: /api/errors/9775bf6455e85424. Report an issue: GitHub.