Significant-Gravitas/AutoGPT · warning · HTTPException

str(exc)

Error message

str(exc)

What it means

HTTP 400 from the credit-transaction export endpoint; admin_export_user_history raised ValueError and its message becomes the response detail. By design the export is capped at CREDIT_EXPORT_MAX_DAYS days and CREDIT_EXPORT_MAX_ROWS rows, and any violation (window too long, or a constraint on the filters) surfaces here as a 400 instead of silently truncated data.

Source

Thrown at autogpt_platform/backend/backend/api/features/admin/credit_admin_routes.py:159

    logger.info(
        "Admin %s exporting credit transactions [%s..%s] type=%s user=%s incl_inactive=%s",
        admin_user_id,
        start.isoformat(),
        end.isoformat(),
        transaction_type.value if transaction_type else None,
        user_id,
        include_inactive,
    )
    try:
        history = await admin_export_user_history(
            start=start,
            end=end,
            transaction_type=transaction_type,
            user_id=user_id,
            include_inactive=include_inactive,
        )
    except ValueError as exc:
        raise HTTPException(status_code=400, detail=str(exc)) from exc
    return CreditTransactionsExportResponse(
        transactions=history,
        total_rows=len(history),
        window_days=(end - start).days,
        max_window_days=CREDIT_EXPORT_MAX_DAYS,
    )


class CopilotUsageExportResponse(BaseModel):
    rows: list[CopilotWeeklyUsageRow]
    total_rows: int
    window_days: int
    max_window_days: int


@router.get(
    "/copilot-usage/export",
    response_model=CopilotUsageExportResponse,

View on GitHub (pinned to 9c8bb5550f)

Solutions

1. Read the detail message to identify which cap was breached.
2. Split the export into consecutive windows of at most CREDIT_EXPORT_MAX_DAYS days (the response reports window_days and max_window_days).
3. Add user_id or transaction_type filters to keep row count under CREDIT_EXPORT_MAX_ROWS.
4. Verify start <= end.

Example fix

# before: one 180-day request
curl '/admin/credits/export?start=2026-01-01T00:00:00Z&end=2026-06-30T00:00:00Z'

# after: two capped windows
curl '/admin/credits/export?start=2026-01-01T00:00:00Z&end=2026-03-31T23:59:59Z'
curl '/admin/credits/export?start=2026-04-01T00:00:00Z&end=2026-06-30T23:59:59Z'

const MAX_DAYS = 90; // keep in sync with CREDIT_EXPORT_MAX_DAYS
function splitWindows(start: Date, end: Date): [Date, Date][] { /* consecutive <=MAX_DAYS slices */ }

try { await exportCredits({ start, end }); } catch (e) { if (e.status === 400 && /days/.test(e.detail)) { for (const [s, t] of splitWindows(start, end)) await exportCredits({ start: s, end: t }); } }

When it happens

Trigger: Requesting a window longer than CREDIT_EXPORT_MAX_DAYS; start after end; or an invalid combination of transaction_type/user_id filters that the data layer rejects.

Common situations: Finance wanting a full-quarter export when the cap is 90 days; admins paginating by widening the range instead of narrowing it; typo'd transaction_type value.

Related errors

• start and end query params are required
• start and end query params are required
• str(exc)
• str(exc)
• Either user_id or email query parameter is required.

AI-assisted analysis of Significant-Gravitas/AutoGPT@9c8bb5550f (2026-08-14). Data as JSON: /api/errors/cb7898c81a6154cc. Report an issue: GitHub.