abhigyanpatwari/GitNexus · critical · SandboxError

transcript artifact must be a regular non-symlink file: {pat

Error message

transcript artifact must be a regular non-symlink file: {path}

What it means

Raised by `_bound_transcript_artifact` when the transcript file is a symlink or not a regular file (directory, fifo, socket, device). Symlinks are blocked because they can escape the results root.

Source

Thrown at eval/workflow_bench/evolve.py:394

            relative, _, _ = _transcript_artifact_metadata(artifact)
            normalized = _normalized_transcript_artifact_path(relative)
            if normalized in seen_paths:
                raise SandboxError(f"duplicate transcript artifact path: {normalized}")
            seen_paths.add(normalized)
        artifacts_by_row.append(artifacts)
    return artifacts_by_row


def _bound_transcript_artifact(root: Path, metadata: Any) -> str:
    relative, expected_digest, expected_size = _transcript_artifact_metadata(metadata)

    path = _results_artifact_path(root, relative, transcript=True)
    try:
        before = path.lstat()
    except OSError as exc:
        raise SandboxError(f"transcript artifact is unavailable: {path}: {exc}") from exc
    if stat.S_ISLNK(before.st_mode) or not stat.S_ISREG(before.st_mode):
        raise SandboxError(f"transcript artifact must be a regular non-symlink file: {path}")
    if stat.S_IMODE(before.st_mode) & 0o077:
        raise SandboxError(f"transcript artifact must be owner-only: {path}")
    if before.st_size != expected_size:
        raise SandboxError(f"transcript artifact size does not match its results row: {path}")

    descriptor = os.open(path, os.O_RDONLY | getattr(os, "O_NOFOLLOW", 0))
    try:
        opened = os.fstat(descriptor)
        if not stat.S_ISREG(opened.st_mode) or opened.st_dev != before.st_dev or opened.st_ino != before.st_ino:
            raise SandboxError(f"transcript artifact changed while opening: {path}")
        digest = hashlib.sha256()
        content = bytearray()
        while chunk := os.read(descriptor, 64 * 1024):
            digest.update(chunk)
            content.extend(chunk)
            if len(content) > MAX_EVIDENCE_FILE_BYTES:
                del content[: len(content) - MAX_EVIDENCE_FILE_BYTES]
        after = os.fstat(descriptor)

View on GitHub (pinned to d540b00184)

Solutions

1. Replace the symlink/irregular node with a real, owned regular file containing the transcript bytes.
2. Re-run the benchmark so the runner writes genuine transcript files.
3. If this appears unexpectedly, treat it as a potential integrity violation and regenerate the results from a trusted source.

import stat
for r in evidence:
    for a in r.get('transcript_artifacts', []):
        p = results_dir / a['path']
        m = p.lstat().st_mode
        if stat.S_ISLNK(m) or not stat.S_ISREG(m):
            raise ValueError(f'transcript is not a regular file: {p}')

When it happens

Trigger: A file at `transcripts/<name>` that is a symbolic link, or whose lstat mode is not S_ISREG (e.g. a directory was created where a transcript file is expected).

Common situations: Someone symlinked a transcript to shared storage, a broken restore left a directory in place of a file, or an adversarial/tampered results dir tries to redirect reads.

Related errors

• evidence source must be a regular non-symlink file: {path}
• transcript artifact must be owner-only: {path}
• {label} must be a real non-symlink directory: {path}
• {label} must be a regular non-symlink file: {path}
• results directory must not traverse symlinks: {root}

AI-assisted analysis of abhigyanpatwari/GitNexus@d540b00184 (2026-08-12). Data as JSON: /api/errors/33ac5f5ccaeac5f1. Report an issue: GitHub.