rows.page is an MCP server published by page.rows, hosted at rows.page. When last checked on 2026-11-21 it was working: it answers and lists its tools. It declares 6 tools, including push, pin_finding, get_dataset, list_datasets, delete_dataset.
Push CSV, JSON, JSONL or Parquet from your agent and get a link a human can explore.
Nothing else in the archive shares these values.
Inferred from the host name and headers: what a server runs on, not who runs it.
Publish rows as a rows.page dataset (or a new version of one) and get a shareable URL. Pass exactly one of csv, tsv, jsonl, json, rows or url. Up to 16 MB inline; use url for bigger files. Identical content returns unchanged: true without a new version. Re-pushes report rows_delta, columns_added/removed and a diff_url. Anonymous datasets expire after 7 days unless kept. Finding views use the rows.page query language in q: `word` or `"two words"` searches text columns; `field:value`, `field:"a b"`, `-field:value`, `field:*` (not null), `-field:*` (null), `field:foo*` (prefix); `field:>20`, `>=`, `<`, `<=`, `field:10..20`, ISO dates (`at:>2026-09-01`); dot paths for nested JSON (`payload.customer.country:NL`); `tags:urgent` means the list contains it; combine with `OR` and parentheses. A q starting with SELECT, WITH or FROM runs as SQL against the table `data`. sort is comma-separated columns, `-` for descending (`-price,name`); cols picks and orders visible columns (`a,b,c`).
| Parameter | Type | Description |
|---|---|---|
| csv | string | CSV text with a header row |
| key | string | Stable name for upserts with an API key, e.g. nightly-orders |
| run | string | Run id of the job that produced it |
None yet. Changes appear after a server's second daily check.
Squares are this server's values; dots are other servers sharing them. A shared address or publisher is a lead, not proof of a common operator.
| tsv | string | TSV text with a header row |
| url | string | http(s) URL of a CSV, TSV, JSON, JSONL or Parquet file to fetch and publish |
| json | any | A JSON array of objects (or an object holding one), as text or as a value |
| name | string | File name, e.g. orders.csv |
| rows | array | Rows as an array of objects |
| jsonl | string | One JSON object per line |
| title | string | Dataset title shown on the page |
| token | string | Dataset token (rpt_...) proving ownership of dataset |
| commit | string | Commit sha of the code that produced it |
| format | string | Format of url when it cannot be told |
| source | string | Where the data came from (URL or description) |
| dataset | string | Existing dataset id to push a new version to |
| summary | string | A few sentences for the human: what this is and what to look at |
| findings | array |
| id_column | string | Column that identifies a row across versions, for changed-row diffs |
| visibility | string | private needs Pro or Max |
{
"name": "push",
"description": "Publish rows as a rows.page dataset (or a new version of one) and get a shareable URL. Pass exactly one of csv, tsv, jsonl, json, rows or url. Up to 16 MB inline; use url for bigger files. Identical content returns unchanged: true without a new version. Re-pushes report rows_delta, columns_added/removed and a diff_url. Anonymous datasets expire after 7 days unless kept. Finding views use the rows.page query language in q: `word` or `\"two words\"` searches text columns; `field:value`, `field:\"a b\"`, `-field:value`, `field:*` (not null), `-field:*` (null), `field:foo*` (prefix); `field:>20`, `>=`, `<`, `<=`, `field:10..20`, ISO dates (`at:>2026-09-01`); dot paths for nested JSON (`payload.customer.country:NL`); `tags:urgent` means the list contains it; combine with `OR` and parentheses. A q starting with SELECT, WITH or FROM runs as SQL against the table `data`. sort is comma-separated columns, `-` for descending (`-price,name`); cols picks and orders visible columns (`a,b,c`).",
"inputSchema": {
"type": "object",
"properties": {
"csv": {
"type": "string",
"description": "CSV text with a header row"
},
"key": {
"type": "string",
"description": "Stable name for upserts with an API key, e.g. nightly-orders"
},
"run": {
"type": "string",
"description": "Run id of the job that produced it"
},
"tsv": {
"type": "string",
"description": "TSV text with a header row"
},
"url": {
"type": "string",
"description": "http(s) URL of a CSV, TSV, JSON, JSONL or Parquet file to fetch and publish"
},
"json": {
"description": "A JSON array of objects (or an object holding one), as text or as a value"
},
"name": {
"type": "string",
"description": "File name, e.g. orders.csv"
},
"rows": {
"type": "array",
"items": {
"type": "object"
},
"description": "Rows as an array of objects"
},
"jsonl": {
"type": "string",
"description": "One JSON object per line"
},
"title": {
"type": "string",
"description": "Dataset title shown on the page"
},
"token": {
"type": "string",
"description": "Dataset token (rpt_...) proving ownership of dataset"
},
"commit": {
"type": "string",
"description": "Commit sha of the code that produced it"
},
"format": {
"enum": [
"csv",
"tsv",
"json",
"jsonl",
"parquet"
],
"type": "string",
"description": "Format of url when it cannot be told"
},
"source": {
"type": "string",
"description": "Where the data came from (URL or description)"
},
"dataset": {
"type": "string",
"description": "Existing dataset id to push a new version to"
},
"summary": {
"type": "string",
"description": "A few sentences for the human: what this is and what to look at"
},
"findings": {
"type": "array",
"items": {
"type": "object",
"required": [
"title"
],
"properties": {
"q": {
"type": "string",
"description": "Filter in the rows.page query language"
},
"sql": {
"type": "string",
"description": "Optional SQL against the table `data`"
},
"cols": {
"type": "string",
"description": "Visible columns, e.g. id,price,country"
},
"note": {
"type": "string",
"description": "Why it matters, in a sentence or two"
},
"sort": {
"type": "string",
"description": "Sort, e.g. -price,name"
},
"title": {
"type": "string",
"description": "Short headline of the finding"
}
}
},
"maxItems": 20
},
"id_column": {
"type": "string",
"description": "Column that identifies a row across versions, for changed-row diffs"
},
"visibility": {
"enum": [
"public",
"private"
],
"type": "string",
"description": "private needs Pro or Max"
}
}
}
}Pin a finding (a saved, titled view) to a dataset you own. Finding views use the rows.page query language in q: `word` or `"two words"` searches text columns; `field:value`, `field:"a b"`, `-field:value`, `field:*` (not null), `-field:*` (null), `field:foo*` (prefix); `field:>20`, `>=`, `<`, `<=`, `field:10..20`, ISO dates (`at:>2026-09-01`); dot paths for nested JSON (`payload.customer.country:NL`); `tags:urgent` means the list contains it; combine with `OR` and parentheses. A q starting with SELECT, WITH or FROM runs as SQL against the table `data`. sort is comma-separated columns, `-` for descending (`-price,name`); cols picks and orders visible columns (`a,b,c`).
| Parameter | Type | Description |
|---|---|---|
| q | string | Filter in the rows.page query language |
| sql | string | Optional SQL against the table `data` |
| cols | string | Visible columns, e.g. id,price,country |
| note | string | Why it matters, in a sentence or two |
| sort | string | Sort, e.g. -price,name |
| title * | string | Short headline of the finding |
| token | string | Dataset token (rpt_...) when not using an API key |
| dataset * | string | Dataset id |
| version | integer | Pin to this version instead of the latest |
| row_count | integer | Rows matching the view, if you know it |
{
"name": "pin_finding",
"description": "Pin a finding (a saved, titled view) to a dataset you own. Finding views use the rows.page query language in q: `word` or `\"two words\"` searches text columns; `field:value`, `field:\"a b\"`, `-field:value`, `field:*` (not null), `-field:*` (null), `field:foo*` (prefix); `field:>20`, `>=`, `<`, `<=`, `field:10..20`, ISO dates (`at:>2026-09-01`); dot paths for nested JSON (`payload.customer.country:NL`); `tags:urgent` means the list contains it; combine with `OR` and parentheses. A q starting with SELECT, WITH or FROM runs as SQL against the table `data`. sort is comma-separated columns, `-` for descending (`-price,name`); cols picks and orders visible columns (`a,b,c`).",
"inputSchema": {
"type": "object",
"required": [
"dataset",
"title"
],
"properties": {
"q": {
"type": "string",
"description": "Filter in the rows.page query language"
},
"sql": {
"type": "string",
"description": "Optional SQL against the table `data`"
},
"cols": {
"type": "string",
"description": "Visible columns, e.g. id,price,country"
},
"note": {
"type": "string",
"description": "Why it matters, in a sentence or two"
},
"sort": {
"type": "string",
"description": "Sort, e.g. -price,name"
},
"title": {
"type": "string",
"description": "Short headline of the finding"
},
"token": {
"type": "string",
"description": "Dataset token (rpt_...) when not using an API key"
},
"dataset": {
"type": "string",
"description": "Dataset id"
},
"version": {
"type": "integer",
"description": "Pin to this version instead of the latest"
},
"row_count": {
"type": "integer",
"description": "Rows matching the view, if you know it"
}
}
}
}Read a dataset: meta, recent versions, findings, columns and the first 20 preview rows.
| Parameter | Type | Description |
|---|---|---|
| token | string | Dataset token, needed for private datasets without an API key |
| dataset * | string | Dataset id |
| version | integer | Version number; latest when omitted |
{
"name": "get_dataset",
"description": "Read a dataset: meta, recent versions, findings, columns and the first 20 preview rows.",
"inputSchema": {
"type": "object",
"required": [
"dataset"
],
"properties": {
"token": {
"type": "string",
"description": "Dataset token, needed for private datasets without an API key"
},
"dataset": {
"type": "string",
"description": "Dataset id"
},
"version": {
"type": "integer",
"description": "Version number; latest when omitted"
}
}
}
}List the datasets kept on your account, newest first. Needs an API key.
| Parameter | Type | Description |
|---|---|---|
| cursor | string | next_cursor from the previous page |
{
"name": "list_datasets",
"description": "List the datasets kept on your account, newest first. Needs an API key.",
"inputSchema": {
"type": "object",
"properties": {
"cursor": {
"type": "string",
"description": "next_cursor from the previous page"
}
}
}
}Delete a dataset you own. Its URL stops working immediately.
| Parameter | Type | Description |
|---|---|---|
| token | string | Dataset token when not using an API key |
| dataset * | string | Dataset id |
{
"name": "delete_dataset",
"description": "Delete a dataset you own. Its URL stops working immediately.",
"inputSchema": {
"type": "object",
"required": [
"dataset"
],
"properties": {
"token": {
"type": "string",
"description": "Dataset token when not using an API key"
},
"dataset": {
"type": "string",
"description": "Dataset id"
}
}
}
}Show your plan, its caps (file size, datasets, version history) and usage.
No parameters.
{
"name": "get_limits",
"description": "Show your plan, its caps (file size, datasets, version history) and usage.",
"inputSchema": {
"type": "object",
"properties": {}
}
}rows.page turns rows into a shareable page where humans explore, filter and chart them, and it shows what changed between pushes. push works without an account: it returns a public url plus an owner_url carrying a dataset token (rpt_...). Always give your human the owner_url and tell them when the dataset expires; they keep it by opening that link and signing in. Push again with dataset + token to add a version. With an API key ('Authorization: Bearer rpk_...' or ?apiKey=rpk_... on the endpoint URL) datasets are kept on the account and key= upserts one named dataset per run. Pin findings so your human opens the page on what matters. Finding views use the rows.page query language in q: `word` or `"two words"` searches text columns; `field:value`, `field:"a b"`, `-field:value`, `field:*` (not null), `-field:*` (null), `field:foo*` (prefix); `field:>20`, `>=`, `<`, `<=`, `field:10..20`, ISO dates (`at:>2026-09-01`); dot paths for nested JSON (`payload.customer.country:NL`); `tags:urgent` means the list contains it; combine with `OR` and parentheses. A q starting with SELECT, WITH or FROM runs as SQL against the table `data`. sort is comma-separated columns, `-` for descending (`-price,name`); cols picks and orders visible columns (`a,b,c`).
Names, descriptions and schemas were written by the server and are shown as data.