A request routes itself to the right team
The message is read and sent where it will be answered: sales, support or accounts.
Public chat · Nothing needed · 12 nodes
Example scenarios
Ready-made scenarios: copy the nodes onto your own canvas or open one as a new draft.
For business
Pulling fields out of text
Turns the free text of a request or an email into fields, so they can be stored, counted or passed on.
Public chat · Nothing needed · 7 nodes
An agent answer in your own format
The model's answer goes through your template, so every answer has the same shape.
Public chat · Nothing needed · 7 nodes
Paste a table into the chat, get a summary
Copy a chunk out of a spreadsheet into a message and get the read straight back: what it is, what stands out, what looks wrong.
Public chat · Nothing needed · 6 nodes
A recording of a stand-up becomes minutes with tasks
Audio becomes structure: what was discussed, what was decided, who does what, and what is still open.
Public chat · Nothing needed · 9 nodes
Draft, critique, rewrite
The model writes, then looks for weaknesses in its own text, then rewrites with them in hand.
Public chat · Nothing needed · 9 nodes
Three angles on a decision, then a verdict
Arguments for, against and a list of risks are worked out in parallel, and the conclusion draws on all three.
Public chat · Nothing needed · 10 nodes
A digest across a batch of documents
Each text is condensed on its own, and only the condensed versions are drawn together.
Public chat · Nothing needed · 10 nodes
One scenario answering in every channel
A shared beginning and separate endings: the question and the answer are one, the formatting is per channel.
Public chat, Telegram, VK · Nothing needed · 12 nodes
A form request, parsed and filed
The form on your site posts a request: the scenario breaks it into fields, writes a row to a table and answers the form straight away.
Webhook · a collection under Tables · 8 nodes
Ask your own table in plain words
The export sits in Files, the question is an ordinary sentence, and the answer is counted from that export's rows.
Public chat · an export file on the Files tab · 9 nodes
Every conversation lands in a table as a summary
The bot answers the customer and records what the conversation was about and how it ended.
Public chat · a collection under Tables · 14 nodes
Reviews sorted by tone and topic
An unhappy review reaches the quality chat immediately, a good one is filed for statistics, and a disguised question goes to support.
Webhook · a collection under Tables and a Telegram bot · 13 nodes
An order that lives in a table between runs
Every message about the order appends a row, and the bot answers knowing its whole history.
Public chat · a collection under Tables · 10 nodes
A sales report arrives in Telegram every morning
Data from a table, a schedule and a messenger in one scenario: yesterday's total, order count and average.
Schedule · a collection of sales and a Telegram bot · 9 nodes
A support bot that answers from your documents
Answers customers from your policies and price list, and does not invent what it could not find. The basis of a first-line bot.
Public chat · a price list or a policy in PDF · 8 nodes
Search across your documents
Finds the right fragment in your documents and passes it on, with no retelling by the model.
Public chat · a price list or a policy in PDF · 6 nodes
The bot calls a human when it cannot cope
Answers from your documents, and when the answer is not there it hands over honestly and tells the customer so.
Public chat · company documents and a Telegram bot for the manager · 14 nodes
What your documents are missing collects itself
The bot answers from your documents, and questions it could not answer land in a list of gaps.
Public chat · a knowledge base and a collection under Tables · 15 nodes
A precise answer from the price list and the policy
Two searches, a fusion and a re-rank: the answer is assembled from both bases and rests on what actually fits.
Public chat · two knowledge bases: a price list and policies · 14 nodes
The bot on your site can see the cart
The scenario asks the page what the visitor has in their cart, and answers with that in mind.
Widget · one function next to the widget snippet · 7 nodes
The site went down, so a message lands in Telegram
Checks every half hour whether the site answers, and writes to you in Telegram only when it has gone quiet.
Schedule · a Telegram bot and an address to check · 6 nodes
The customer taps instead of typing
A menu of buttons instead of a conversation: a tap re-enters the same scenario and takes the right branch.
Telegram · a Telegram bot · 10 nodes
A chat message is written into the amoCRM deal
A customer writes about their deal and the scenario updates its card, then confirms what was recorded.
Public chat · an amoCRM account with API access · 11 nodes
Any service can now write to you in Telegram
A three-step bridge: another system calls an address, you read the event in your messenger.
Webhook · a Telegram bot · 7 nodes
Someone joined the group and the bot greets them first
Greets everyone who joins by name and points them at the rules straight away.
Telegram · a Telegram bot added to the group as an administrator · 7 nodes
A VK community menu
Three buttons instead of a conversation, with an allowance for clients that do not render buttons.
VK · a VK community with messages enabled · 12 nodes
A morning email about what landed overnight
One email instead of forty: what arrived, from whom, and which of it wants an answer.
Schedule · a mailbox with IMAP access · 9 nodes
A list of only the emails waiting on you
Not a retelling of the inbox but a filter: the three emails where somebody wants something, not the forty.
Schedule · a mailbox with IMAP access · 9 nodes
An urgent email reaches Telegram within the minute
The mailbox is checked often, but you only hear from it when something will not wait until tomorrow.
Schedule · a mailbox and a Telegram bot · 11 nodes
A request from the inbox becomes a row in a table
An email from a customer turns into a row with a name, an address and the gist; newsletters pass by.
Schedule · a mailbox and a collection under Tables · 12 nodes
An email from a customer, an answer from your documents
The same document search the support bot uses, except the question arrives by email and the answer leaves by email.
Schedule · a mailbox and a knowledge base with your documents · 13 nodes
“What's on today”, a list of meetings
The bot answers with a fact from your calendar rather than with something the model remembers.
Public chat · a calendar with CalDAV access · 9 nodes
The bot tells a customer when you are free
A calendar reports busy time; the scenario works out the free slots itself and names them by the hour.
Public chat · a calendar with CalDAV access · 8 nodes
A customer books an appointment in the conversation
A request in plain words becomes an event in your calendar.
Public chat · a calendar with CalDAV write access · 10 nodes
A reminder an hour before, and exactly one
The schedule fires often, the reminder arrives once: the scenario remembers what it has already said.
Schedule · a CalDAV calendar, a Telegram bot and a collection under Tables · 14 nodes
A secretary you ask in ordinary words
One scenario answers both «what's on today» and «when am I free on Thursday»: the model decides what to ask the calendar.
Public chat · a calendar with CalDAV access · 9 nodes
For yourself
Describe it in words, get a picture
A short request is expanded into a detailed brief, and the picture is drawn from that.
Telegram · Nothing needed · 8 nodes
Mark the habit, the bot keeps the streak
Type «done», get the count of days in a row. Not one model call, so a run costs almost nothing.
Telegram · Nothing needed · 9 nodes
A photo of a receipt becomes a line of spending
Photograph the receipt, send it to the bot: the amount, the category and the place are already in a table.
Telegram · a collection under Tables · 9 nodes
“How much did I spend”, a breakdown by category
Ask in plain words, and the bot counts from your own records and answers in sums.
Telegram · a collection of spending records · 7 nodes
A shopping list the bot keeps for you
Write «get more milk» in plain words; the bot adds, shows and clears the list on request.
Telegram · a collection under Tables · 16 nodes
“Remind me on Friday”, and on Friday it does
One scenario with two doors: a person sets a reminder in chat, the schedule delivers the ones that came due.
Telegram, Schedule · a collection under Tables and a Telegram bot · 14 nodes
A second brain: throw thoughts in, ask in your words
Everything you write to the bot lands in a knowledge base, and a month later you can ask about it in a normal sentence.
Telegram · an empty knowledge base (the scenario fills it) · 13 nodes
Send a link, get a summary that stays in your base
The bot opens the page, summarises it and files the summary in your knowledge base.
Telegram · an empty knowledge base (the scenario fills it) · 9 nodes
Talk it out, get a tidy note
Send a voice message, get back text worth rereading, with any tasks pulled out into a list.
Telegram · a Telegram bot · 7 nodes
A voice message in, a voice answer out
The full voice loop: heard, thought about, answered aloud. Your hands stay busy with something else.
Telegram · a Telegram bot · 9 nodes
Technical
A plain chat with a model
The shortest scenario there is: a question comes in, the model answers, the answer goes out. A good place to start.
Public chat · Nothing needed · 6 nodes
A chat that remembers the conversation
Remembers what was said earlier, so the customer does not have to repeat themselves every message.
Public chat · Nothing needed · 7 nodes
An agent that searches the web
Answers from current information on the web and shows its sources when the answer is not in your documents.
Public chat · Nothing needed · 7 nodes
A long text edited in pieces
The document is split on paragraph boundaries, each piece is edited on its own, and it is glued back in order.
Public chat · Nothing needed · 12 nodes
The agent calls your own scenario as a tool
A slice of graph that queries your table becomes a tool the model calls when it decides to.
Public chat · a collection of stock levels · 10 nodes
A check that notices more than an outage
Asks the service how it feels and tells three outcomes apart: down, answering wrongly, healthy.
Schedule · a Telegram bot and an address to check · 10 nodes
One summary for every address
Checks a list of addresses in turn and sends a single message with one line per address.
Schedule · a Telegram bot and a list of addresses · 13 nodes
Primary is down, try the backup; both down, call a human
A cascade of two sources with a human called at the very end, when the automation is out of options.
Schedule · a Telegram bot and two source addresses · 10 nodes