These docs cover the S6S workflow engine — the developer playground layer underneath the product. For the main AI Visibility product (brand monitoring, the unified Report, fix actions), start at /docs/ai-visibility.

Node Types

Every workflow is a series of steps connected in a flow. Each step has an action type that determines what it does. This is the comprehensive reference for all built-in node types.

Triggers

3 node types

Actions

21 node types

Logic

5 node types

Dynamic Integrations

Triggers

Triggers define how a workflow starts. Every workflow needs exactly one trigger.

`trigger.webhook`

Start a workflow when an HTTP request hits a specific path.

Config	Type	Default	Description
path	string	/my-webhook	URL path (e.g. /new-lead)
method	string	POST	HTTP method to accept

Output:{ body, headers, query, method, path, receivedAt }

Invoke URL: https://s6s.ai/api/webhooks/{path}?token={secret}

Access trigger data in your workflow steps:

{{trigger.output.body.*}}
{{trigger.output.headers.*}}
{{trigger.output.query.*}}
{{trigger.output.method}}

`trigger.schedule`

Run a workflow on a recurring schedule using cron expressions.

Config	Type	Default	Description
cronExpression	string	0 * * * *	Cron expression (minute, hour, day, month, weekday)
timezone	string	UTC	IANA timezone name

Output:{ scheduledTime, timezone }

Common cron patterns:

*/5 * * * * — every 5 minutes
0 9 * * 1-5 — weekdays at 9 AM
0 0 * * * — daily at midnight
0 0 1 * * — first of every month

`trigger.manual`

Start a workflow manually via the UI "Run" button or the POST /api/workflows/{id}/run API. No configuration needed.

Output:{ payload, triggeredAt, triggeredBy }

payload — the input object from the run request body (or {} if none)

triggeredAt — ISO timestamp

triggeredBy — "api" or "ui"

Recipe format

{
  "trigger": { "type": "manual" }
}

Actions

Actions do work — call APIs, send messages, transform data, run code.

`action.http`

Make an HTTP request to any URL.

Config	Type	Default	Description
url	string	—	Full URL
method	string	GET	HTTP method
headers	object	{}	Request headers
body	string	—	Request body (for POST/PUT/PATCH)

Output:{ status, headers, body }

Supports expression templates in all fields: {{steps.prev.output.url}}

`action.transform`

Reshape data between steps using a JSON template with expressions.

Config	Type	Default	Description
template	object	{}	JSON template with {{expression}} placeholders

Output: The evaluated template object.

Example

{
  "template": {
    "message": "Hello, {{trigger.output.body.name}}!",
    "timestamp": "{{trigger.output.timestamp}}"
  }
}

`action.code`

Execute a JavaScript expression and return the result.

Config	Type	Default	Description
expression	string	({ result: 'hello' })	JavaScript expression to evaluate
language	string	javascript	Language (currently only javascript)

Output: The return value of the expression.

`action.email`

Send an email via SMTP.

Config	Type	Default	Description
to	string	—	Recipient email
subject	string	—	Email subject
bodyTemplate	string	—	Email body (supports expressions)

Requires credential: smtp

`action.slack`

Post a message to a Slack channel.

Config	Type	Default	Description
channel	string	#general	Channel name or ID
textTemplate	string	—	Message text (supports expressions)

Requires credential: slack

`action.discord`

Send a message or perform an action in Discord.

Config	Type	Default	Description
action	string	send_message	send_message, send_embed, add_reaction
channelId	string	—	Discord channel ID
content	string	—	Message content

Requires credential: discord

`action.telegram`

Send a message or perform an action in Telegram.

Config	Type	Default	Description
action	string	send_message	send_message, send_photo, send_document
chatId	string	—	Telegram chat ID
text	string	—	Message text

Requires credential: telegram

Auto-chunking: Messages longer than 4096 characters are automatically split at newline boundaries and sent as multiple messages.

`action.imap`

Fetch, search, or read mailbox content over IMAP.

Config	Type	Default	Description
action	string	fetch_recent	fetch_recent, search, read_message
query	string	—	Search query for search mode
messageId	string	—	Message identifier for read_message mode
maxResults	number	20	Maximum messages to return
since	string	—	Relative time filter like 1h or 7d
folder	string	INBOX	Mailbox folder
markAsRead	boolean	false	Mark fetched messages as read

Requires credential: imap

`action.rss`

Fetch and parse an RSS feed.

Config	Type	Default	Description
feedUrl	string	—	RSS feed URL
maxItems	number	10	Maximum items to return

Output:{ items: [{ title, link, description, pubDate }] }

`action.twitter`

Post a tweet or interact with the Twitter API.

Config	Type	Default	Description
action	string	post_tweet	post_tweet, search, get_user
text	string	—	Tweet content

Requires credential: twitter

`action.google_sheets`

Read from or write to Google Sheets.

Config	Type	Default	Description
action	string	read_rows	read_rows, append_row, update_cell
spreadsheetId	string	—	Spreadsheet ID from the URL
range	string	Sheet1!A:Z	Cell range

Requires credential: google

`action.google_drive`

List, upload, or download files from Google Drive.

Config	Type	Default	Description
action	string	list	list, upload, download, create_folder
maxResults	number	10	Max files to return

Requires credential: google

`action.gmail`

Search, read, or send email via the Gmail API.

Config	Type	Default	Description
action	string	search_messages	search_messages, read_message, send_message
query	string	—	Gmail search query (for search_messages)
messageId	string	—	Message ID (for read_message)
maxResults	number	10	Max results for search
to	string	—	Recipient (for send_message)
subject	string	—	Email subject (for send_message)
body	string	—	Email body (for send_message)

Requires credential: google

`action.database`

Execute a SQL query against a connected database.

Config	Type	Default	Description
query	string	—	SQL query
operation	string	select	select, insert, update, delete

Requires credential: database

`action.filter`

Filter or transform arrays.

Config	Type	Default	Description
arrayExpression	string	—	Expression that resolves to an array
condition	string	true	Filter condition
mode	string	filter	filter or map

`action.ai`

Send a prompt to an LLM and get a response.

Config	Type	Default	Description
prompt	string	—	The prompt (supports expressions)
responseFormat	string	text	text or json

`action.buffer`

Full Buffer publishing pipeline: create/edit/delete posts, view queue & sent history, and publish immediately across X, Instagram, TikTok, and more.

Config	Type	Default	Description
action	string	create_post	create_post, list_channels, get_queue, get_sent, share_now, delete_post, update_post
text	string	—	Post text (create_post, update_post)
mediaUrl	string	—	URL of media to attach
channelIds	array	[]	Buffer channel IDs to publish to (create_post)
scheduledAt	string	—	ISO timestamp to schedule the post
shorten	boolean	true	Shorten links in post text
profileId	string	—	Profile/channel ID (get_queue, get_sent)
updateId	string	—	Buffer update ID (share_now, delete_post, update_post)
count	number	20	Results per page (get_queue, get_sent)
page	number	—	Page number for pagination
since	string	—	Unix timestamp — only return updates after this time

Requires credential: buffer

`action.webhook_response`

Return a custom HTTP response to the webhook caller.

Config	Type	Default	Description
statusCode	number	200	HTTP status code
body	string	—	Response body

`action.execute_workflow`

Trigger another S6S workflow by slug.

Config	Type	Default	Description
slug	string	—	Target workflow slug
waitForCompletion	boolean	true	Wait for the child workflow to finish

`action.stop_and_error`

Immediately stop the workflow and report an error. Useful for guardrails.

Config	Type	Default	Description
message	string	Workflow stopped	Error message surfaced in the run summary

`action.youtube`

Upload videos, list channel videos, get video details, and update video metadata via the YouTube Data API.

Config	Type	Default	Description
action	string	list_videos	upload_video, list_videos, get_video, update_video
videoUrl	string	—	URL of the video to upload
title	string	—	Video title
description	string	—	Video description
tags	array	[]	Video tags for discovery
privacyStatus	string	private	public, unlisted, or private
categoryId	string	—	YouTube video category ID
videoId	string	—	Video ID (for get_video / update_video)
maxResults	number	10	Max videos to return (for list_videos)

Requires credential: google

Logic

Logic nodes control execution flow — branching, looping, timing, and parallelism.

`logic.condition`

Branch execution based on a condition. Has two output handles: true (green) and false (red).

Config	Type	Default	Description
expression	string	—	Left-hand value (supports expressions)
operator	string	eq	Comparison operator (see below)
value	string	—	Right-hand value

Operators:

eqneqgtgteltltecontainsexistsnot_existsmatchesin

`logic.switch`

Route to different branches based on an expression's value.

Config	Type	Default	Description
expression	string	—	Value to switch on
cases	array	[]	[{ value, target }] case definitions
defaultTarget	string	—	Fallback target if no case matches

`logic.loop`

Iterate over an array, executing downstream nodes for each item.

Config	Type	Default	Description
arrayExpression	string	—	Expression resolving to an array
itemVariable	string	item	Variable name for current item
maxIterations	number	100	Safety limit to prevent infinite loops

`logic.wait`

Pause execution for a specified duration.

Config	Type	Default	Description
seconds	number	60	Wait time in seconds

`logic.merge`

Combine outputs from parallel branches into a single object. Place after parallel groups in the flow array.

Config	Type	Default	Description
mode	string	wait_all	Merge strategy (currently wait_all)

Output:{ branch_a: {...}, branch_b: {...} }

Recipe example

{
  "flow": [
    "fetch_data",
    ["branch_a", "branch_b"],
    "combine",
    "send_result"
  ]
}

// "combine" is a merge step — it waits for both
// parallel branches and merges their outputs.

Conditional Execution

Any node can include a run_if expression in its config. When present, the engine evaluates it at runtime: truthy executes the node normally, falsy skips it with status skipped. Skipped nodes do not fail the run.

run_if example

{
  "ref": "send_alert",
  "action": "telegram",
  "config": {
    "action": "send_message",
    "chatId": "123456",
    "text": "Alert: threshold exceeded",
    "run_if": "{{steps.check.output.amount > 1000}}"
  }
}

Dynamic Integrations

Beyond the built-in action types, S6S supports a provider.action syntax for any API integration. Use this to connect to services like Notion, Pinterest, HubSpot, or any other API — S6S learns the API automatically.

Action	Description
notion.create_page	Create a page in Notion
pinterest.create_pin	Create a pin on Pinterest
hubspot.create_contact	Create a contact in HubSpot
stripe.create_invoice	Create a Stripe invoice

S6S resolves the provider and action, fetches the API schema, and handles authentication, pagination, and error retries. Learn more about dynamic integrations.

Expressions

Use {{expression}} syntax in any config field to reference data from previous steps:

Expression	Description
{{trigger.output.body.name}}	Webhook request body field
{{steps.format.output.message}}	A previous step's output by ref name
{{steps.fetch.output.body.items[0]}}	Nested array access
{{run.id}}	Current run ID

Expressions are evaluated at runtime with full access to the workflow execution context.

Need help choosing node types?

Describe what you want to automate in the chat UI and S6S will generate the right workflow for you.

Start Building Back to Docs