Function Commands

Commands for managing function registries and viewing function execution logs.

function

Manage functions

Usage

tailor function [command]

See Global Options for options available to all commands.

Commands

Command	Aliases	Description
function get	-	Get a function registry by name
function list	-	List function registries in a workspace
function logs	-	List or get function execution logs.
function run	test-run	Run a function on the Tailor Platform server without deploying.
function script	-	Scaffold a one-off script to run with function run.

function get

Get a function registry by name

Usage

tailor function get [options]

Options

Option	Alias	Description	Required	Default	Env
--workspace-id <WORKSPACE_ID>	-w	Workspace ID	No	-	TAILOR_PLATFORM_WORKSPACE_ID
--profile <PROFILE>	-p	Workspace profile	No	-	TAILOR_PLATFORM_PROFILE
--name <NAME>	-n	Function name	Yes	-	-

See Global Options for options available to all commands.

function list

List function registries in a workspace

Usage

tailor function list [options]

Options

Option	Alias	Description	Required	Default	Env
--workspace-id <WORKSPACE_ID>	-w	Workspace ID	No	-	TAILOR_PLATFORM_WORKSPACE_ID
--profile <PROFILE>	-p	Workspace profile	No	-	TAILOR_PLATFORM_PROFILE
--order <ORDER>	-	Sort order (asc or desc)	No	"desc"	-
--limit <LIMIT>	-l	Maximum number of items to return (0 or omit: unlimited)	No	-	-

See Global Options for options available to all commands.

function logs

List or get function execution logs.

Usage

tailor function logs [options] [execution-id]

Arguments

Argument	Description	Required
execution-id	Execution ID (if provided, shows details with logs)	No

Options

Option	Alias	Description	Required	Default	Env
--workspace-id <WORKSPACE_ID>	-w	Workspace ID	No	-	TAILOR_PLATFORM_WORKSPACE_ID
--profile <PROFILE>	-p	Workspace profile	No	-	TAILOR_PLATFORM_PROFILE
--order <ORDER>	-	Sort order (asc or desc)	No	"desc"	-
--limit <LIMIT>	-l	Maximum number of items to return (0: unlimited)	No	50	-
--follow	-f	Keep polling a running execution and print new log entries as they arrive (detail mode only)	No	false	-
--interval <INTERVAL>	-i	Polling interval for --follow (e.g., '3s', '500ms', '1m')	No	"3s"	-
--timeout <TIMEOUT>	-t	Maximum time to keep following (e.g., '30s', '10m'); unbounded by default	No	-	-

See Global Options for options available to all commands.

Examples

List all function execution logs

$ tailor function logs

Get execution details with logs

$ tailor function logs <execution-id>

Output as JSON

$ tailor function logs --json

Get execution details as JSON

$ tailor function logs <execution-id> --json

Stream log entries of a running execution until it completes

$ tailor function logs <execution-id> --follow

Notes

Each execution includes errorKind, which tells you where a failure came from: USER_RUNTIME (your code threw while running, or exceeded its time, memory, or CPU limit), USER_NON_RUNTIME (your script failed while loading, before your function was called, such as when its top-level code threw), PLATFORM (a failure on the Platform side, or the Platform aborting the run because of its own deadline or cancellation), NONE (no error), or UNSPECIFIED (not recorded, or a kind this CLI version does not recognize). A workflow job waiting on another job also reports USER_NON_RUNTIME when that job fails or is canceled. errorName and errorMessage hold the error's name and message, or null when none was recorded. The list table omits errorMessage; use --json or the execution details to read it.

Execution details include logEntries, the structured log lines (message, severity, timestamp) recorded while the function ran. They are available while the execution is still running. The logs string joins their messages with newlines.

Use --follow to keep polling a running execution and print new log entries as they arrive until it completes. Polling continues while the execution is suspended at a wait point, and indefinitely unless --timeout is set. With --json, --follow waits for completion and then emits the final execution details once.

When viewing a specific execution that failed, the command displays error details with the stack trace mapped back to your original source files (clickable file links and code snippets, matching function run output).

Stack traces are mapped only when the execution includes a content hash for the exact build that ran. If the content hash is missing or the build is no longer available, the command falls back to a plain-text error display.

function run

Run a function on the Tailor Platform server without deploying.

Aliases: test-run

Usage

tailor function run [options] <file>

Arguments

Argument	Description	Required
file	Path to the function file	Yes

Options

Option	Alias	Description	Required	Default	Env
--workspace-id <WORKSPACE_ID>	-w	Workspace ID	No	-	TAILOR_PLATFORM_WORKSPACE_ID
--profile <PROFILE>	-p	Workspace profile	No	-	TAILOR_PLATFORM_PROFILE
--name <NAME>	-n	Workflow job name to run (matches the name field of createWorkflowJob)	No	-	-
--arg <ARG>	-a	JSON argument to pass to the function	No	-	-
--machine-user <MACHINE_USER>	-m	Machine user name for authentication. Falls back to the active profile's default machine user.	No	-	TAILOR_PLATFORM_MACHINE_USER_NAME
--config <CONFIG>	-c	Path to SDK config file	No	"tailor.config.ts"	-
--allow-schema-drift	-	Run a scaffolded script even when its schema snapshot no longer matches	No	false	-

See Global Options for options available to all commands.

Examples

Run a resolver with input arguments

$ tailor function run resolvers/add.ts --arg '{"a":1,"b":2}'

Run a specific workflow job by name

$ tailor function run workflows/sample.ts --name validate-order

Run a pre-bundled .js file directly

$ tailor function run build/resolvers/add.js --arg '{"a":1,"b":2}'

Notes

You can pass either a source file (.ts) or a pre-bundled file (.js). When a .js file is provided, detection and bundling are skipped and the file is executed as-is.

A script scaffolded by function script with a generated db.ts is checked against its db.snapshot.json before execution and refused on schema drift; pass --allow-schema-drift to run it anyway. The check compares table and field structure; hook and validator code changes are not detected.

test-run is a deprecated alias of this command and will be removed in v3.

WARNING

Workflow job .start() calls do not work in this mode. Started jobs are not executed; only the target job's body function runs in isolation.

function script

Scaffold a one-off script to run with function run.

Usage

tailor function script [options] <file>

Arguments

Argument	Description	Required
file	Path to create the script at (must end with .ts)	Yes

Options

Option	Alias	Description	Required	Default	Env
--workspace-id <WORKSPACE_ID>	-w	Workspace ID	No	-	TAILOR_PLATFORM_WORKSPACE_ID
--profile <PROFILE>	-p	Workspace profile	No	-	TAILOR_PLATFORM_PROFILE
--config <CONFIG>	-c	Path to Tailor config file	No	"tailor.config.ts"	TAILOR_CONFIG_PATH
--namespace <NAMESPACE>	-	Target TailorDB namespace (required when the config does not pin one)	No	-	-
--remote	-	Generate script-scoped DB types from the deployed schema	No	false	-

See Global Options for options available to all commands.

Examples

Scaffold a one-off script (single-namespace project)

$ tailor function script scripts/fix-prices.ts

Scaffold a script targeting a specific namespace

$ tailor function script scripts/fix-prices.ts --namespace tailordb

Scaffold from a deployed or external namespace

$ tailor function script scripts/fix-prices.ts --namespace shared --remote

Notes

The scaffolded script is a plain default-exported function; execute it with tailor function run <file>.

By default, when the project configures kyselyTypePlugin, the skeleton imports getDB() from the plugin's generated types. Without the plugin, the command uses the namespace's local table definitions to write a script-scoped db.ts plus a db.snapshot.json next to the script; function run refuses to run the script when that snapshot no longer matches the deployed or locally defined table and field structure.

Pass --remote to generate the script-scoped files from the deployed schema instead, even when kyselyTypePlugin is configured. This is required for an external namespace. Re-running the command refreshes db.ts and db.snapshot.json from the selected source and leaves the script itself untouched.