ScripteraAI Infrastructure
414 scenarios
Your first AI agent with toolsVery quick start: your first scenario in minutesThe basics in 3 easy stepsConnect a service with no ready-made integration
Knowledge Base Skeleton (RAG)
- An AI chat that answers from a source's fresh data
- Ask your own documents — get answers from them
- Keep your assistant's knowledge base always up to date
- A chat assistant over your PDF documents
- A chat assistant over a large rulebook or code
- A knowledge base from your documents where every piece keeps its context
- An assistant that answers from all your sources at once
- A chatbot that answers from your documents
- Search your own images — by what's shown in them
- Documents from storage — into a base for AI search
- An assistant that answers from the files in your storage
- Knowledge-base pages into a foundation for AI search
- A support bot in Telegram that answers from your docs and learns
- A chat assistant over your knowledge base with adaptive answers
- Chat with your documents — answers from their content
- A chat over your documents with answers from the source
- A chat over your documents from a folder
- An assistant that remembers your files and answers from them
- How to build an assistant that answers from your own documents
- A CV assistant for your portfolio site
- Search inside videos by asking in plain words
- Sync your Notion pages into an AI knowledge base
- Questions and answers over your own documents
- Cloud documents — into a knowledge base for AI
- Ask your Google Drive documents in plain words
- Answers from your documents in plain words
- Answers from your documents with a source reference
- A chatbot over your site — answering from its own content
- Document search by meaning and by exact words
- Precise answers from your own documents
- A chatbot that answers from your documents
- Questions to a PDF document — answers straight from its text
- A knowledge base for an AI from website data
- Precise answers from your knowledge base
- Answers from your library of texts — with quotes
- Questions across all your documents — one precise answer
- An AI knowledge base from a spreadsheet of questions and answers
- A whole site — into a text base for a neural net
- Answers from a big document — just ask in plain words
- A website chatbot grounded in your documents
- An assistant that answers from your documents
- Loading files into a database for meaning-based search
- Chat with your folder of documents — even scans
- A chatbot that answers from your documents
- A document assistant with a Pinecone knowledge base
- A chatbot that answers from your documents and updates itself
- A knowledge base from your cloud files — with answers
- Answers from your documents right in the chat
- Questions about chart-heavy documents — answers straight from the pages
- Answers to questions about your own documents
- An answer from your documents, the news and academic papers
- Search and answers over a collection of texts
- Smart document search that keeps itself up to date
- A study assistant that answers from your own materials
- A knowledge base from PDF brochures — for instant AI search
- Search across your PDFs: text recognition and smart answers
- Answers from your PDFs right in Telegram
- Answers from your own documents — just ask
- A chatbot that answers from your documents
- A chat assistant answering from your knowledge base, with memory
- A chat over your documents — a two-step answer
- Answers from a long document — with a page reference
- A smart assistant over your own spreadsheet data
- A chat assistant over documents from a folder
- Question-answer over your library of texts
- Chatbot from your site content and client data
- A knowledge base over your PDFs — that answers questions
- A meaning-search base over your video transcripts
- Answers from your own documents, in plain words
- A knowledge base from web pages — one link at a time
- An assistant knowledge base that stays in step with your documents
- A folder of documents — turned into a knowledge base for your assistant
- A chat assistant over your own documents
- A knowledge base from websites and chat answers — from one link
- A chatbot knowledge base from a folder of documents
- Voice notes into text and a knowledge base — automatically
- A knowledge base from audio with Q&A — automatically
- Ask your own PDFs and get answers from them
- A topic map built from your document folder
- A catalogue helper: answers buyers' questions from your own products
- A bot that answers questions from your documents
- A bot that answers about your company and captures leads — automatically
- Answers from your documents with a source — automatically
- Support email replies from your own knowledge base
- Questions to your own documents — answered from the text, not from memory
- A new document in the folder — instantly ready for questions
- Meaning-based search over a medical-procedure catalog
- A YouTube channel as a searchable knowledge base — automatically
- The assistant's knowledge updates with the file — and the old version is deleted
- Ask your documents questions, answered from them — in chat
- A knowledge base from emails and notes, answered in chat
- A ticket assistant answering from their history — in chat
- A folder of documents — turned into a map of topic connections
- Support ticket triage — simple ones close, hard ones go to an engineer
- A knowledge base built from your document folder: ask, get an answer
- Support replies from your knowledge base — as a draft in Gmail
- A knowledge base that updates itself — from a storage folder
- A chat assistant with memory and answers from your documents
- A bot that answers from your own documents
- A chatbot over your knowledge base — answers from documents, no making things up
- Meaning-based document search — from Oracle Database
- Document analysis and recommendations — right in your team chat
- A site chatbot that answers from your knowledge base
- Customer email replies — grounded in your knowledge base
- An assistant that answers from your cloud documents
- Support email replies from your knowledge base
- A whole website into a knowledge base — for search and answers from its content
- A knowledge base for your chatbot from any website
- Your site into the bot's memory — refreshed daily
- Questions about your documents — from any channel
- A smart chat over your documents with a daily summary
- Documents into crisp meaning-blocks for precise helper answers
- A Telegram bot that answers your customers' common questions
- Tickets answered from your own knowledge base — as a comment, immediately
- A Telegram bot that answers from your own documents
- Customer emails answered from your knowledge base
- A support chatbot from your website — answering from your materials
- A work-chat assistant that remembers your context
- Support replies from your own knowledge base
- Chat with your documents from a cloud folder
- Questions about your document — right in the chat
- Matching scenario templates by the meaning of your request
- A support bot trained on your own website
- A chat assistant that answers from your documents
- Your own site as a knowledge base for a chatbot
- A knowledge base from Jira issues — searchable by meaning
Building Blocks & Utilities
- Connect a service with no ready-made integration
- Turn a recording into text, plus ready AI examples
- Auto-post and update messages in your team chat
- Sending an email — as a scenario step
- Publishing a series of posts in a batch — from a list
- One combined list instead of scattered pieces
- A bot that hears you out and answers all at once
- An email with an attachment — assembled and sent for you
- Personalized emails and pages from a template — assembled for you
- A form with a dropdown that updates itself
- A messenger bot that replies with a picture
- Put text on an image on request
- Image details from a link — format, size, weight
- Fetch the latest Hacker News articles in one step
- Paced sending with a wait-for-approval step
- Auto-install the libraries your server needs
- Gather data from several sources into one list
- SMS blast to a contact list from a table
- Passing an access key from a form — a worked example
- A messenger weather bot that replies with a chart
- Saving data to a file on disk
- A ready report from your records, emailed
- Process a big list without hitting service limits
- Scheduled database routines with a chat report
- Batch-run tasks from a text file
- Counting the number of records a step returned
- AI prompts pulled from a repo, no scenario edits
- Trigger a scenario on a new email in Email Trigger (IMAP)
- Strapi entries — create, update, verify
- Automatic record handling in Bubble
- Posting a file to a messenger channel
- Fresh website data to your messenger — on schedule
- Merging a burst of messages into one before processing
- Pack files into an archive and upload to the cloud
- One scenario for requests from many sources
- A chat assistant that remembers past conversations
- A welcome for every new hire in the team chat — automatically
- The neural net's answer always in the right format
- Splitting messenger messages into text and files
- Large table exports without rate-limit errors
- Aggregate a folder of documents into one dictionary
- Export all contacts from HubSpot
- Export a board from Monday.com
- Register bot users into a spreadsheet
- Approvals in your messenger — one button
- Different assistant permissions for different users
- Memory for scenarios: a value by key in a file
- A meaning-based cache for AI answers — lower costs
- Guard against overlapping scheduled runs
- A Telegram bot that hears you out
- Message your messenger from any program in one request
- Channel and member setup in Slack — automatically
- Rich bot replies in Telegram — with media and no cut-offs
- Smart message buffering for an AI chat
- A chatbot that remembers clients, backed by Supabase
- Secure links for clients to connect their socials
- Manage room members in Matrix
- File-based key-value store: read by key
- A chat that runs commands on your server
- A ready preview for any link
- An access layer for your bot — insiders only
- A connection under watch — it drops, it recovers itself
- Auto-update your Twitter profile banner
- A lock on the job — no double runs
- Concurrency lock to prevent double runs — a ready block
- Self-updating images and links in materials — automatically
- Marked-up text from your sheet, converted to web-ready markup
- A small list app — with no separate server
- Approvals with one tap in a chat — verified and logged
- Your calendar in words, not clicks
- Coordinates — turned into address, weather, timezone and sun times
- Shared memory for your scenarios: never fetch the same answer twice
- A short text summary through a form — straight into a sheet
- Marked-up text — turned into a properly formatted knowledge-base page
- Auto-renewing social page access
- A single work log for your scenarios
- Duplicate-processing guard with a shared 'done' list
- Scenario breakpoints with a button in Slack
MCP Server
- Your own scenarios as tools for an AI agent
- Run your calendar with words in a chat
- A personal assistant: meetings, emails, and a sheet — from one phrase
- Your own assistant over your service: secure data access on command
- Memory and database access for your AI assistant
- A calendar you run in words — with no clashes
- Your email as a toolkit for an AI agent
- AI-assistant access to your database
- An AI agent that works with your Airtable base
- Personal money tracking by voice — expenses and budget in a sheet
- Assistant access to Google Drive files — as a tool
- Manage your spreadsheet with AI — read, edit, create sheets
- An AI assistant that tidies your storage
- One assistant for mail, calendar, files and social
- Manage your calendar from a chat
- How to build an assistant that answers from your own documents
- An AI assistant for issues and comments in GitHub
- An AI assistant runs your Google Calendar
- Smart review-base search for your AI assistant
- Ad campaign data on request from an AI assistant
- Manage your Supabase database through an AI agent
- All of Pipedrive CRM as tools for an AI agent
- Run your calendar in plain words
- Email control from an AI assistant
- Inbox management from an AI assistant: labels and status
- Manage your Odoo CRM in plain language
- Manage a WordPress site through an AI assistant
- Manage your Pipedrive CRM from a messenger via an AI assistant
- Control Reddit from your AI assistant — posts, comments, subreddits
- Managing Jira tasks — by plain conversation
- Jira control for your AI assistants
- Calls and SMS driven by AI
- SharePoint access for your AI assistant — every operation
- A community bot you drive with words
- Hacker News access for your AI assistant
- Bubble app data for an AI agent
- Managing your monitoring dashboards through a helper
- Managing your workouts through an assistant
- End-of-day market data for your AI assistant
- Giving your AI assistant access to your work apps
- Full Google Drive access for your AI assistant — all 17 operations ready
- Site certificates — as a tool your assistant can use
- Your to-do list, run by an AI assistant
- An MCP server for Gong — a ready set of operations for an AI agent
- Run your store from a chat — in plain words
- Manage CRM leads in plain language
- Managing KlickTipp contacts by assistant command
- Accounting on request: invoices, payments and customers in plain words
- On-call incidents, run through an AI assistant
- Transcription on request, handled by your assistant
- An assistant that researches, then writes the email
- Full support-desk reach for your AI helper
- Run your Trello board by talking to an assistant
- Full AI-assistant access to your code storage — 18 ready actions
- Full SyncroMSP access for your AI assistant
- Full code-project control for an AI assistant
- Web reading and search for your AI assistant
- Manage CircleCI builds from your assistant
- Work with documents from your assistant
- A chat assistant for your QuickBooks accounting
- An AI assistant with direct access to your database
- An AI assistant that runs Spotify for you
- Controlling Gotify notifications from an AI agent
- Ready NASA data access for your AI assistant
Demos & Examples
- A routine assistant: email, document answers and meeting booking
- A showcase of possibilities: what scenarios are built from
- Search your photos — by a plain-words description
- A personal email assistant — handles your mail and attachments
- Get a sticker's ID in a messenger — in one message
- Suggest dishes by a guest's taste
- Extracting data from documents into a table
- A top-creators ranking in a ready report
- Find objects in a photo by description — AI-powered
- A form from your table — built automatically
- ISS coordinates every minute to a device channel
- Logging data to a database on a schedule — automatically
- Downloading a file from a site by its button — automatically
- An AI assistant you tune in conversation
- On-demand images from a chat description
- The space station's real-time position in your base
- An assistant over your PDFs, with answer-quality checks
- A new channel with members and an intro message — in one run
- Campaigns to your list — created and sent automatically
- Check your messenger server connection: list every channel and category
- Create and update an incident in PagerDuty
- Contacts in your mailer — created and updated automatically
- Check a repository's community profile in one request
- Create a customer and add them straight to a mailing segment
- Database records: created, updated and verified — with no manual entry
- A stream of sensor data into a queue — for testing and simulation
- New newsletter subscribers — straight into a table
- A list of all your support mailboxes
- Add a record to your table and read it back
- Contacts from a table — straight into outreach
- Your email-service contacts, kept up to date automatically
- An outreach campaign: created, filled with contacts, verified
- Contacts in your client base — created, completed, verified
- An instant signal on every sale — automatically
- Kick off a project build in one move
- The book you find lands on your shelf right away
- Bulk-update contacts that match a condition
- Contacts in your mailing base create and update themselves
- Webinars created and updated automatically
- Members and their activity — from a sheet into your community platform
- Your email-service subscribers, kept up to date automatically
- New subscribers land in your table on their own
- A community member set up in a single run
- A new team channel: created, described, and started
- Your marketing-tool contacts, kept up to date automatically
- Check that an address is real before you email it
- Creating a client in your tracking system — in one step
- A person's profile from a single email
- A company in your CRM — created automatically
- A new project in Copper — an instant trigger for any scenario
- Create a billing customer without typing it in
- Send an event to your analytics — from any scenario
Tutorials & Starters
- Your first AI agent with tools
- Very quick start: your first scenario in minutes
- The basics in 3 easy steps
- Your first AI assistant with tools — step by step
- An expert bot for your documents — answers to the point
- AI fills in descriptions in your Google Sheet
- An assistant that answers clients and books appointments
- An illustrated article to your site from one topic — on its own
- A personal chat assistant — calendar, weather, news, email
- A short weekly meeting summary from your calendar
- Route records by rules — a ready building block
- Example: joining data across steps
- Secure request-based access to your Supabase data
- Automatic record-keeping in a table
- Data analysis in a spreadsheet: merge, filter, summarise
- Transform data and upload the file to storage
- Auto-export a table to a file in storage
- Shaping and cleaning data inside a scenario — step by step
- Image generation and cloud storage — automatically
- Checking a helper's answers against reference examples
- AI images from a description — straight into your cloud
- An assistant's reply with your approval before sending
- One scenario — three ways to start, one shared logic
Local AI on Your Server
- Ask your PDFs questions, get answers right in chat
- Answers from your own documents — just ask in plain words
- Text-to-speech voiced on your own server
- An assistant that remembers your files and answers from them
- Your personal AI assistant in messaging — on your own server
- Questions and answers over your own documents
- Smart search over your own email — in plain questions
- Chat with your documents — ask and get an answer
- Legal document analysis with a ready report
- Voice-over for your texts — in a batch
- A chat assistant over your own documents
- Paid access to your AI model — pay per request
- WhatsApp auto-replies and lead capture — automatically
- A short digest of any file in a folder — automatically
- Meeting minutes from a video recording — automatically
- Support auto-replies from your own knowledge base — over email
- Instant email replies, with an alert to your messenger
- A chatbot over your knowledge base — answers from documents, no making things up
- Voice memos from your phone, as text in your notes
- Email answers from your own approved guidelines
Agent Quality Evaluation
- AI assistant check: how correctly it picks its actions
- A/B testing chat assistant versions on live conversations
- Compare several AIs' answers with scoring
- Check the quality of your AI assistant's answers
- Measuring document-search accuracy
- Checking your AI's answer accuracy against a reference
- Checking your AI assistant's quality across several measures
- Scoring your assistant's answers on how grounded they are in your documents
- Testing AI categorization accuracy on a dataset
- Evaluate your AI assistant's answer accuracy
- Compare answers from different AI models in Google Sheets
- Check that your AI assistant actually uses its tool
- Score your AI's accuracy against expected answers
- Relevance scoring for your AI assistant's answers
- Quality scoring for the assistant's summaries
- Scoring your AI assistant's answer quality — automatically
- AI bot chat logs from your database into a spreadsheet
- Checking AI model accuracy on documents — in a table
- Checking a helper's answers against reference examples
- Testing your helper's safeguards: a test suite and an accuracy report
AI Agent with Web Access
- Collecting data from company sites into your database — automatically
- Collect website data into a spreadsheet — the AI looks and enters it
- An assistant that works on websites for you — clicks, types, extracts
- A chat assistant with internet search
- Search the web right from your chat
- A chat assistant with web search
- Answers about a website in a chat
- A bot that gathers website data on request
- An assistant that does browser tasks on a chat command
- Chat answers grounded in fresh facts from the web
