abhigyanpatwari/GitNexus · error · SandboxError

transcript_artifacts exceeds the per-row session limit of {M

Error message

transcript_artifacts exceeds the per-row session limit of {MAX_TRANSCRIPT_ARTIFACTS_PER_ROW}

What it means

Raised by `_preflight_transcript_artifacts` when a single evidence row carries more than `MAX_TRANSCRIPT_ARTIFACTS_PER_ROW` (=2) transcript artifacts. Each row corresponds to one benchmark run, which the runner caps at 2 captured sessions, so >2 is an invariant violation.

Source

Thrown at eval/workflow_bench/evolve.py:369

        or relative.parts[0] != "transcripts"
        or any(part in {"", ".", ".."} for part in relative.parts)
    ):
        raise SandboxError(f"unsafe results artifact path: {relative_value!r}")
    return relative.as_posix()


def _preflight_transcript_artifacts(evidence: list[dict[str, Any]]) -> list[list[Any]]:
    """Bound every transcript reference before any evidence file is read."""

    artifacts_by_row: list[list[Any]] = []
    seen_paths: set[str] = set()
    total = 0
    for artifacts_row in evidence:
        artifacts = artifacts_row.get("transcript_artifacts", [])
        if not isinstance(artifacts, list):
            raise SandboxError("transcript_artifacts must be a list")
        if len(artifacts) > MAX_TRANSCRIPT_ARTIFACTS_PER_ROW:
            raise SandboxError(
                f"transcript_artifacts exceeds the per-row session limit of {MAX_TRANSCRIPT_ARTIFACTS_PER_ROW}"
            )
        total += len(artifacts)
        if total > MAX_TRANSCRIPT_ARTIFACTS:
            raise SandboxError(f"transcript_artifacts exceeds the global evidence limit of {MAX_TRANSCRIPT_ARTIFACTS}")
        for artifact in artifacts:
            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)

View on GitHub (pinned to d540b00184)

Solutions

1. Inspect results.jsonl and locate the row with >2 transcript_artifacts entries.
2. Trim the list to the 2 entries the runner would actually produce, or re-run the benchmark with the stock runner.
3. If you genuinely need more, raise MAX_TRANSCRIPT_ARTIFACTS_PER_ROW in evolve.py and MAX_TRANSCRIPT_ARTIFACTS accordingly and update the tests that pin the cap.

MAX_PER_ROW = 2
violators = [(r.get('task'), len(r.get('transcript_artifacts', []))) for r in evidence if len(r.get('transcript_artifacts', [])) > MAX_PER_ROW]
if violators:
    raise ValueError(f'rows over per-row transcript limit: {violators}')

When it happens

Trigger: `proposer_evidence_entries` is called with an `evidence` list where some row's `transcript_artifacts` list has length 3 or more.

Common situations: A results.jsonl produced by a modified runner that captures extra sessions, or a row assembled by stitching several runs together.

Related errors

• transcript_artifacts exceeds the global evidence limit of {M
• transcript_artifacts must be a list
• duplicate transcript artifact path: {normalized}
• transcript artifact is unavailable: {path}: {exc}
• transcript artifact must be a regular non-symlink file: {pat

AI-assisted analysis of abhigyanpatwari/GitNexus@d540b00184 (2026-08-12). Data as JSON: /api/errors/109d982d5bf45f59. Report an issue: GitHub.