resample_timeseries

Group rows into time buckets: one output row per interval of a timestamp column. The same engine that backs the GUI Time series dialog and CLI --resample. Read-only analytics (stays available under --mcp-read-only).

When to use

• Turn per-transaction rows into daily / weekly / monthly totals.
• Produce one series per category with group_by, ready to chart.

Input schema

Parameter	Type	Required?	Default	Description
path	string	yes*	(no default)	Path to the file (omit when open_tab is set)
open_tab	string	no	(no default)	Operate on an open GUI tab (@active or a tab name)
table	string	no	(no default)	Specific table for multi-table sources
time_col	string	yes	(no default)	The timestamp column to bucket
value_cols	string[]	yes	(no default)	Columns aggregated within each bucket
interval	string	no	day	minute/hour/day/week/month/quarter/year
agg	string	no	sum	sum/mean/min/max/count/first/last
group_by	string[]	no	[]	One series per combination of these columns
limit	integer	no	server default	Cap response rows (0 = unlimited)
unlimited	bool	no	false	Lift the 5,000,000-row file-loader cap so the resample sees every row

The bucket lands in a column named bucket, or bucket_2 if the source already has a column called bucket.

Response shape

The resampled table in the standard row-returning shape:

{
  "schema": [ { "name": "bucket", "type": "Utf8" }, … ],
  "rows": [ [ … ], … ],
  "row_count": 12,
  "truncated": false,
  "total_rows_available": 12,
  "cell_truncated": false
}

Example call

{
  "name": "resample_timeseries",
  "arguments": {
    "path": "/tmp/sales.csv",
    "time_col": "day",
    "value_cols": ["amount"],
    "interval": "month",
    "agg": "sum",
    "group_by": ["region"]
  }
}

Notes

The timestamp column is cast with TRY_CAST, so one unparseable value lands in a null bucket instead of failing the whole call.

See also

• rolling_window: moving averages over the previous N rows.
• pivot: reshape between long and wide form.
• run_sql: custom date_trunc grouping when the parameters do not fit.