Tinybird
App · 24 actions
Work with your Tinybird workspace: add rows, import files, run SQL, call endpoints, run copy and sink pipes, follow jobs, sign read tokens and manage secrets, and start workflows on new rows, jobs and changed endpoint results.
Actions
| Action | What it does |
|---|---|
| Add rows | Add one row, or a list of rows from an earlier step, to a data source, and wait until Tinybird has stored them. |
| Cancel copy jobs | Cancel a copy pipe's waiting and running jobs and pause its schedule. |
| Cancel job | Cancel a job that is still waiting (or a populate that is running). |
| Create read token | Sign a token that may only read one endpoint (optionally with parameters fixed, such as one customer's id) or one data source, and stops working at the time given. |
| Create secret | Store a value, such as a database password, that pipes read with tb_secret('name') without showing it. |
| Delete rows | Delete the rows of a data source that match a SQL condition, as a job, or only count them first. |
| Delete secret | Delete a secret. |
| Empty data source | Remove every row from a data source, keeping the data source and its columns. |
| Get data source | Get one data source: its columns and their types, how many rows it holds, its engine and the pipes that read it. |
| Get endpoint results | Call a published endpoint with values for its parameters and return its rows. |
| Get job | Get one job's status, timings, rows and error, for 48 hours after it was created. |
| Get pipe | Get one pipe: what kind it is, the parameters it takes, its nodes' SQL and, for a copy or sink pipe, its schedule. |
| Import file from URL | Have Tinybird load a CSV, NDJSON or Parquet file from a web address into a data source, adding its rows or replacing the data. |
| List data sources | List the workspace's data sources (its tables) with their columns, size and the pipes that read them. |
| List jobs | List the workspace's jobs (imports, copies, sinks, deletes) of the last 48 hours, at most 100, newest first. |
| List pipes | List the workspace's pipes (its endpoints, copy, sink and other pipes) with the parameters each takes. |
| List secrets | List the workspace's secrets (the values pipes read with tb_secret) by name. |
| List tokens | List the workspace's tokens by name with what each may do. |
| Pause copy schedule | Stop a scheduled copy pipe from starting new runs. |
| Resume copy schedule | Let a paused copy pipe's schedule start runs again. |
| Run copy pipe | Run a copy pipe now, writing its result into its target data source. |
| Run sink pipe | Run a sink pipe now, exporting its result to its bucket or Kafka topic. |
| Run SQL query | Run a SQL SELECT over the workspace's data sources and pipes and return the rows. |
| Update secret | Give an existing secret a new value. |
Add rows
Add one row, or a list of rows from an earlier step, to a data source, and wait until Tinybird has stored them. Fails, saying why, if a row doesn't match the data source's columns.
Inputs
- Data source
data_sourcetextRequired - The data source. Pick one, or give its name.
- Row
rowobject - A value for each column. Numbers and true or false are converted to the column's type.
- Rows
rowslist - A list of rows from an earlier step, each an object of column names and values. Added after the row above.
Outputs
data_sourcetext- The data source the rows went to
quarantined_rowsnumber- How many rows were set aside: always 0, since any fails the step
rows_addednumber- How many rows Tinybird stored
rows_sentnumber- How many rows were sent
Example output
{
"data_source": "events",
"quarantined_rows": 0,
"rows_added": 2,
"rows_sent": 2
}Cancel copy jobs
Cancel a copy pipe's waiting and running jobs and pause its schedule. 'Resume copy schedule' starts it again.
Inputs
- Copy pipe
pipetextRequired - The copy pipe. Pick one, or give its name.
Outputs
cancelled_jobstrue/false- False when no job was waiting or running
pipetext- The copy pipe
schedule_crontext- Its schedule, in cron format
schedule_statustext- Whether the schedule is running or paused
Example output
{
"cancelled_jobs": true,
"pipe": "copy_daily_totals",
"schedule_cron": "0 * * * *",
"schedule_status": "paused"
}Cancel job
Cancel a job that is still waiting (or a populate that is running). Rows it already wrote stay.
Inputs
- Job ID
job_idtextRequired - The job's id, from a step that started it or 'List jobs'.
Outputs
jobobject- The job, now cancelling
Example output
{
"job": {
"created_at": "2026-11-02 10:41:07.398102",
"data_source": "daily_totals",
"error": "",
"id": "c8ae13ef-e739-40b6-8bd5-b1e07c8671c2",
"is_cancellable": false,
"job_url": "https://api.tinybird.co/v0/jobs/c8ae13ef-e739-40b6-8bd5-b1e07c8671c2",
"kind": "copy",
"pipe": "copy_daily_totals",
"row_count": null,
"rows_affected": null,
"started_at": "",
"status": "cancelling",
"updated_at": "2026-11-02 10:41:07.398128"
}
}Create read token
Sign a token that may only read one endpoint (optionally with parameters fixed, such as one customer's id) or one data source, and stops working at the time given.
Inputs
- Token name
nametextRequired - A name for the token, shown in Tinybird's logs.
- Reads
resource_typetext - What the token may read. Defaults to an endpoint.One of:
pipe(An endpoint (pipe)),data_source(A data source). - Endpoint
pipetext - The endpoint the token may read.Only when Reads is An endpoint (pipe).
- Data source
data_sourcetext - The data source the token may read.Only when Reads is A data source.
- Fixed parameters
fixed_paramsobject - Parameter values the token always uses, which whoever holds it can't change, such as a customer's id.Only when Reads is An endpoint (pipe).
- Expires
expires_atdate and timeRequired - When the token stops working.
Outputs
expires_atnumber- When it expires, in Unix seconds
nametext- The token's name
scopetext- What it may read, such as PIPES:READ top_users
tokentext- The token, to send as a Bearer token or ?token=
urltext- For an endpoint, an address that returns its rows (NDJSON) with the token; empty for a data source
Example output
{
"expires_at": 1793599082,
"name": "customer_dashboard",
"scope": "PIPES:READ top_users",
"token": "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.example.signature",
"url": "https://api.tinybird.co/v0/pipes/top_users.ndjson?token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.example.signature"
}Create secret
Store a value, such as a database password, that pipes read with tb_secret('name') without showing it.
Inputs
- Name
nametextRequired - The secret's name: letters, digits and _, unique in the workspace.
- Value
valuetextRequired - The secret's value, at most 8 KB. Tinybird never shows it again.
Outputs
secretobject- The secret
Example output
{
"secret": {
"created_at": "2026-11-02T10:27:57",
"edited_by": "token: 'workspace admin token'",
"name": "pg_password",
"type": "secret",
"updated_at": "2026-11-02T10:29:57"
}
}Delete rows
Delete the rows of a data source that match a SQL condition, as a job, or only count them first. Materialized views fed by it keep their rows.
Inputs
- Data source
data_sourcetextRequired - The data source. Pick one, or give its name.
- Delete rows where
conditiontextRequired - A SQL condition on the data source's columns, such as country = 'ES' or timestamp < '2026-01-01'.
- Only count them
dry_runtrue/false - Count the matching rows without deleting anything.
Outputs
data_sourcetext- The data source
dry_runtrue/false- True when the rows were only counted
jobobject- The delete job
job_idtext- The delete job's id, empty when only counted
rows_to_deletenumber- How many rows match, when only counted
statustext- The delete job's status, empty when only counted
Example output
{
"data_source": "events",
"dry_run": false,
"job": {
"created_at": "2026-11-02 10:41:07.398102",
"data_source": "events",
"error": "",
"id": "c8ae13ef-e739-40b6-8bd5-b1e07c8671c2",
"is_cancellable": true,
"job_url": "https://api.tinybird.co/v0/jobs/c8ae13ef-e739-40b6-8bd5-b1e07c8671c2",
"kind": "delete_data",
"pipe": "",
"row_count": null,
"rows_affected": null,
"started_at": "",
"status": "waiting",
"updated_at": "2026-11-02 10:41:07.398128"
},
"job_id": "c8ae13ef-e739-40b6-8bd5-b1e07c8671c2",
"rows_to_delete": null,
"status": "waiting"
}Delete secret
Delete a secret. Pipes that read it stop working.
Inputs
- Secret
nametextRequired - The secret. Pick one, or give its name.
Outputs
deletedtrue/false- Always true once deleted
nametext- The secret deleted
Example output
{
"deleted": true,
"name": "pg_password"
}Empty data source
Remove every row from a data source, keeping the data source and its columns. This can't be undone.
Inputs
- Data source
data_sourcetextRequired - The data source. Pick one, or give its name.
Outputs
data_sourcetext- The data source emptied
emptiedtrue/false- Always true once emptied
Example output
{
"data_source": "events",
"emptied": true
}Get data source
Get one data source: its columns and their types, how many rows it holds, its engine and the pipes that read it.
Inputs
- Data source
data_sourcetextRequired - The data source. Pick one, or give its name.
Outputs
data_sourceobject- The data source
Example output
{
"data_source": {
"bytes": 430833,
"columns": [
{
"default_value": null,
"name": "timestamp",
"nullable": false,
"type": "DateTime64(3)"
},
{
"default_value": null,
"name": "user_id",
"nullable": false,
"type": "String"
},
{
"default_value": null,
"name": "amount",
"nullable": false,
"type": "Float64"
}
],
"created_at": "2026-10-01 09:00:00.000000",
"description": "Events from the web app.",
"engine": "MergeTree",
"format": "ndjson",
"id": "t_bd1c62b5e67142bd9bf9a7f113a2b6ea",
"name": "events",
"row_count": 3980,
"sorting_key": "timestamp, user_id",
"updated_at": "2026-10-28 13:11:19.464343",
"used_by": [
"top_users"
]
}
}Get endpoint results
Call a published endpoint with values for its parameters and return its rows.
Inputs
- Endpoint
pipetextRequired - The endpoint. Pick one, or give its name.
- Parameters
parametersobject - A value for each parameter the pipe takes. Leave one empty for its default. A list is sent as comma-separated values.
Outputs
bytes_readnumber- How many bytes were read
columnslist- The columns, in order
elapsed_secondsnumber- How long it took, in seconds
pipetext- The endpoint called
row_countnumber- How many rows came back
rowslist- The rows, one object per row
rows_before_limitnumber- At least how many rows matched before any LIMIT
rows_readnumber- How many rows were read
Example output
{
"bytes_read": 95520,
"columns": [
{
"name": "user_id",
"type": "String"
},
{
"name": "total",
"type": "Float64"
}
],
"elapsed_seconds": 0.0042,
"pipe": "top_users",
"row_count": 1,
"rows": [
{
"total": 120.5,
"user_id": "u_1842"
}
],
"rows_before_limit": 1,
"rows_read": 3980
}Get job
Get one job's status, timings, rows and error, for 48 hours after it was created.
Inputs
- Job ID
job_idtextRequired - The job's id, from a step that started it or 'List jobs'.
Outputs
jobobject- The job
Example output
{
"job": {
"created_at": "2026-11-02 10:41:07.398102",
"data_source": "daily_totals",
"error": "",
"id": "c8ae13ef-e739-40b6-8bd5-b1e07c8671c2",
"is_cancellable": true,
"job_url": "https://api.tinybird.co/v0/jobs/c8ae13ef-e739-40b6-8bd5-b1e07c8671c2",
"kind": "copy",
"pipe": "copy_daily_totals",
"row_count": 24,
"rows_affected": null,
"started_at": "",
"status": "done",
"updated_at": "2026-11-02 10:41:07.398128"
}
}Get pipe
Get one pipe: what kind it is, the parameters it takes, its nodes' SQL and, for a copy or sink pipe, its schedule.
Inputs
- Pipe
pipetextRequired - The pipe. Pick one, or give its name.
Outputs
pipeobject- The pipe
Example output
{
"pipe": {
"created_at": "2026-10-01 09:10:00.000000",
"description": "The users who spent most.",
"id": "t_55c39255e6b548dd98cb6da4b7d62c1c",
"name": "top_users",
"nodes": [
{
"id": "t_h65c788b42ce4095a4789c0d6b0156c3",
"name": "endpoint",
"sql": "%\nSELECT user_id, sum(amount) AS total FROM events GROUP BY user_id ORDER BY total DESC LIMIT {{Int32(limit, 10)}}",
"type": "endpoint"
}
],
"parameters": [
{
"default": 10,
"description": "How many users",
"name": "limit",
"required": false,
"type": "Int32"
}
],
"schedule_cron": "",
"schedule_status": "",
"type": "endpoint",
"updated_at": "2026-10-28 13:33:40.850186",
"url": "https://api.tinybird.co/v0/pipes/top_users.json"
}
}Import file from URL
Have Tinybird load a CSV, NDJSON or Parquet file from a web address into a data source, adding its rows or replacing the data. Returns the import job; 'Get job' follows it.
Inputs
- Data source
data_sourcetextRequired - The data source. Pick one, or give its name.
- File address
urltextRequired - The file's web address, which Tinybird downloads.
- File format
file_formattext - The file's format. Defaults to CSV.One of:
csv(CSV (default)),ndjson(NDJSON (one JSON object per line)),parquet. - Existing rows
modetext - Add the file's rows to what is there, or replace the data source's rows with them. Defaults to adding.One of:
append(Add the file's rows (default)),replace(Replace the data with the file's rows). - Replace only rows where
replace_conditiontext - When replacing, a SQL condition such as date >= '2026-10-01': only matching rows are replaced. Leave empty to replace them all.Only when Existing rows is Replace the data with the file's rows.
Outputs
data_sourcetext- The data source the file goes to
jobobject- The job
job_idtext- The id of the job Tinybird started
statustext- The job's status when it was started, usually waiting
Example output
{
"data_source": "events",
"job": {
"created_at": "2026-11-02 10:41:07.398102",
"data_source": "events",
"error": "",
"id": "c8ae13ef-e739-40b6-8bd5-b1e07c8671c2",
"is_cancellable": true,
"job_url": "https://api.tinybird.co/v0/jobs/c8ae13ef-e739-40b6-8bd5-b1e07c8671c2",
"kind": "import",
"pipe": "",
"row_count": null,
"rows_affected": null,
"started_at": "",
"status": "waiting",
"updated_at": "2026-11-02 10:41:07.398128"
},
"job_id": "c8ae13ef-e739-40b6-8bd5-b1e07c8671c2",
"status": "waiting"
}List data sources
List the workspace's data sources (its tables) with their columns, size and the pipes that read them.
Inputs
None.
Outputs
countnumber- How many
data_sourceslist- The data sources
Example output
{
"count": 1,
"data_sources": [
{
"bytes": 430833,
"columns": [
{
"default_value": null,
"name": "timestamp",
"nullable": false,
"type": "DateTime64(3)"
},
{
"default_value": null,
"name": "user_id",
"nullable": false,
"type": "String"
},
{
"default_value": null,
"name": "amount",
"nullable": false,
"type": "Float64"
}
],
"created_at": "2026-10-01 09:00:00.000000",
"description": "Events from the web app.",
"engine": "MergeTree",
"format": "ndjson",
"id": "t_bd1c62b5e67142bd9bf9a7f113a2b6ea",
"name": "events",
"row_count": 3980,
"sorting_key": "timestamp, user_id",
"updated_at": "2026-10-28 13:11:19.464343",
"used_by": [
"top_users"
]
}
]
}List jobs
List the workspace's jobs (imports, copies, sinks, deletes) of the last 48 hours, at most 100, newest first.
Inputs
- Kind
job_kindtext - Only jobs of this kind. Leave empty for all.One of:
import,copy(Copy pipe),sink(Sink pipe),populateview(Materialized view populate),delete_data(Delete rows),deployment. - Status
statustext - Only jobs in this status. Leave empty for all.One of:
waiting,working,done,error(Failed),cancelling,cancelled. - Pipe
pipetext - Only jobs of this pipe. Leave empty for all.
- Created after
created_afterdate and time - Only jobs created after this time.
- Created before
created_beforedate and time - Only jobs created before this time.
Outputs
countnumber- How many
jobslist- The jobs
Example output
{
"count": 1,
"jobs": [
{
"created_at": "2026-11-02 10:41:07.398102",
"data_source": "daily_totals",
"error": "",
"id": "c8ae13ef-e739-40b6-8bd5-b1e07c8671c2",
"is_cancellable": true,
"job_url": "https://api.tinybird.co/v0/jobs/c8ae13ef-e739-40b6-8bd5-b1e07c8671c2",
"kind": "copy",
"pipe": "copy_daily_totals",
"row_count": null,
"rows_affected": null,
"started_at": "",
"status": "done",
"updated_at": "2026-11-02 10:41:07.398128"
}
]
}List pipes
List the workspace's pipes (its endpoints, copy, sink and other pipes) with the parameters each takes.
Inputs
- Kind
pipe_typetext - Only pipes of this kind. Leave empty for all.One of:
endpoint,copy,sink,materialized(Materialized view),default(Plain pipe).
Outputs
countnumber- How many
pipeslist- The pipes
Example output
{
"count": 1,
"pipes": [
{
"created_at": "2026-10-01 09:10:00.000000",
"description": "The users who spent most.",
"id": "t_55c39255e6b548dd98cb6da4b7d62c1c",
"name": "top_users",
"nodes": [
{
"id": "t_h65c788b42ce4095a4789c0d6b0156c3",
"name": "endpoint",
"sql": "%\nSELECT user_id, sum(amount) AS total FROM events GROUP BY user_id ORDER BY total DESC LIMIT {{Int32(limit, 10)}}",
"type": "endpoint"
}
],
"parameters": [
{
"default": 10,
"description": "How many users",
"name": "limit",
"required": false,
"type": "Int32"
}
],
"schedule_cron": "",
"schedule_status": "",
"type": "endpoint",
"updated_at": "2026-10-28 13:33:40.850186",
"url": "https://api.tinybird.co/v0/pipes/top_users.json"
}
]
}List secrets
List the workspace's secrets (the values pipes read with tb_secret) by name. Tinybird never returns their values.
Inputs
None.
Outputs
countnumber- How many
secretslist- The secrets
Example output
{
"count": 1,
"secrets": [
{
"created_at": "2026-11-02T10:27:57",
"edited_by": "token: 'workspace admin token'",
"name": "pg_password",
"type": "secret",
"updated_at": "2026-11-02T10:29:57"
}
]
}List tokens
List the workspace's tokens by name with what each may do. The tokens themselves are left out.
Inputs
None.
Outputs
countnumber- How many
tokenslist- The tokens
Example output
{
"count": 1,
"tokens": [
{
"description": "",
"id": "017a167c-5b1b-46b5-8791-897e87c4d293",
"name": "workspace admin token",
"scopes": [
"ADMIN"
]
}
]
}Pause copy schedule
Stop a scheduled copy pipe from starting new runs. A run already copying finishes.
Inputs
- Copy pipe
pipetextRequired - The copy pipe. Pick one, or give its name.
Outputs
pipetext- The copy pipe
schedule_crontext- Its schedule, in cron format
schedule_statustext- Whether the schedule is running or paused
Example output
{
"pipe": "copy_daily_totals",
"schedule_cron": "0 * * * *",
"schedule_status": "paused"
}Resume copy schedule
Let a paused copy pipe's schedule start runs again.
Inputs
- Copy pipe
pipetextRequired - The copy pipe. Pick one, or give its name.
Outputs
pipetext- The copy pipe
schedule_crontext- Its schedule, in cron format
schedule_statustext- Whether the schedule is running or paused
Example output
{
"pipe": "copy_daily_totals",
"schedule_cron": "0 * * * *",
"schedule_status": "running"
}Run copy pipe
Run a copy pipe now, writing its result into its target data source. Returns the copy job; 'Get job' follows it.
Inputs
- Copy pipe
pipetextRequired - The copy pipe. Pick one, or give its name.
- Parameters
parametersobject - A value for each parameter the pipe takes. Leave one empty for its default. A list is sent as comma-separated values.
- Target's rows
modetext - Add the result to the target data source, or replace its rows with it. Defaults to the pipe's own setting.One of:
append(Add to the target's rows (default)),replace(Replace the target's rows).
Outputs
jobobject- The job
job_idtext- The id of the job Tinybird started
pipetext- The copy pipe
statustext- The job's status when it was started, usually waiting
Example output
{
"job": {
"created_at": "2026-11-02 10:41:07.398102",
"data_source": "daily_totals",
"error": "",
"id": "c8ae13ef-e739-40b6-8bd5-b1e07c8671c2",
"is_cancellable": true,
"job_url": "https://api.tinybird.co/v0/jobs/c8ae13ef-e739-40b6-8bd5-b1e07c8671c2",
"kind": "copy",
"pipe": "copy_daily_totals",
"row_count": null,
"rows_affected": null,
"started_at": "",
"status": "waiting",
"updated_at": "2026-11-02 10:41:07.398128"
},
"job_id": "c8ae13ef-e739-40b6-8bd5-b1e07c8671c2",
"pipe": "copy_daily_totals",
"status": "waiting"
}Run sink pipe
Run a sink pipe now, exporting its result to its bucket or Kafka topic. Returns the sink job; 'Get job' follows it.
Inputs
- Sink pipe
pipetextRequired - The sink pipe. Pick one, or give its name.
- Parameters
parametersobject - A value for each parameter the pipe takes. Leave one empty for its default. A list is sent as comma-separated values.
- File name
file_templatetext - A file name for this run instead of the pipe's own, such as daily_export_{date}.
- File format
file_formattext - A format for this run instead of the pipe's own.One of:
csv,ndjson,parquet. - Compression
compressiontext - Compress the files for this run.One of:
none,gz(Gzip). - When the file exists
write_strategytext - What to do when a file of that name is already there.One of:
new(Add a new file beside an existing one),truncate(Replace an existing file).
Outputs
jobobject- The job
job_idtext- The id of the job Tinybird started
pipetext- The sink pipe
statustext- The job's status when it was started, usually waiting
Example output
{
"job": {
"created_at": "2026-11-02 10:41:07.398102",
"data_source": "",
"error": "",
"id": "c8ae13ef-e739-40b6-8bd5-b1e07c8671c2",
"is_cancellable": true,
"job_url": "https://api.tinybird.co/v0/jobs/c8ae13ef-e739-40b6-8bd5-b1e07c8671c2",
"kind": "sink",
"pipe": "export_daily",
"row_count": null,
"rows_affected": null,
"started_at": "",
"status": "waiting",
"updated_at": "2026-11-02 10:41:07.398128"
},
"job_id": "c8ae13ef-e739-40b6-8bd5-b1e07c8671c2",
"pipe": "export_daily",
"status": "waiting"
}Run SQL query
Run a SQL SELECT over the workspace's data sources and pipes and return the rows. Values can be passed as parameters and read in the query with {{String(name)}}.
Inputs
- SQL
sqltextRequired - The SELECT to run, without a FORMAT clause. A pipe's name reads it like a table.
- Parameters
parametersobject - Values the query reads by name, such as user_id read with {{String(user_id)}}.
Outputs
bytes_readnumber- How many bytes were read
columnslist- The columns, in order
elapsed_secondsnumber- How long it took, in seconds
row_countnumber- How many rows came back
rowslist- The rows, one object per row
rows_before_limitnumber- At least how many rows matched before any LIMIT
rows_readnumber- How many rows were read
Example output
{
"bytes_read": 95520,
"columns": [
{
"name": "user_id",
"type": "String"
},
{
"name": "total",
"type": "Float64"
}
],
"elapsed_seconds": 0.0042,
"row_count": 1,
"rows": [
{
"total": 120.5,
"user_id": "u_1842"
}
],
"rows_before_limit": 1,
"rows_read": 3980
}Update secret
Give an existing secret a new value.
Inputs
- Secret
nametextRequired - The secret. Pick one, or give its name.
- Value
valuetextRequired - The secret's value, at most 8 KB. Tinybird never shows it again.
Outputs
secretobject- The secret
Example output
{
"secret": {
"created_at": "2026-11-02T10:27:57",
"edited_by": "token: 'workspace admin token'",
"name": "pg_password",
"type": "secret",
"updated_at": "2026-11-02T10:29:57"
}
}