Narrated, self-driving tutorials
Paste what you’d say to a colleague. Tuti Tuti works out which sentences click, which ones point, and which ones just talk, then walks your real web app with a human cursor and a narrator, and records the video.
Free is the full product with a one-minute preview. Pro plays it all and exports. Runs in your browser; your recordings never leave it, and your account stays in Canada.
Tuti Tuti - powered by Pandacat.
Your AI writes the script. Tuti Tuti performs it. Pick the branch that matches where you are.
Teach your AI to build an app a script can land on, and let it check its own work.
Every button gets a name a person would say, every field a label, every dialog a role. Plain accessibility, and it is what lets a script land first time. The app guide
Claude Code, Cursor, or any MCP client gets the guides, the parser, and a checker that returns the same warnings the extension would. Set up MCP
Open your app, open the panel, paste, Rehearse. Tuti Tuti walks the screens and says exactly where the words and the labels disagree.
One download after rehearsal, straight to your AI. It fixes the app or the sentence, and the two agree.
Three boxes, no setup. The script comes from the AI you already talk to.
Claude, ChatGPT, Cursor, Lovable, whichever one knows your app. Fill in the screens and the demo values.
Use the tuti-tuti MCP server (https://tutituti.ca/api/mcp) or the script guide at https://tutituti.ca/guide. Write a Tuti Tuti tutorial script for my app: one sentence per beat, exact on-screen labels, open menus and dialogs before referring to what they reveal, a transition sentence at the start of each chapter, blank lines between chapters with a short title line. Screens to cover: [list them]. Demo values to type: [list them]. Check the script with tuti_check_script until it returns no warnings, then give me the plain text.
One sentence per beat, exact on-screen labels, menus opened before what they reveal. The script guide is what it follows.
Open your app, open the Tuti Tuti panel, paste the script, press Rehearse. Fix any red flag by rewording or pointing.
A human cursor, a spotlight on each element, captions synced to the voice. The MP4 lands in your Downloads, blur and all, one file or one per chapter.
One extension, many kinds of video. Each page below answers the questions people ask about that kind.
No recording takes · No editing timeline · Re-render one chapter when the app changes
Real product, not mock-ups · Your words, your voice choice · No studio, no editor
One task each · Captions and spotlight · Twenty short videos from one script
A prompt that works · MCP tools for checking · The rework file closes the loop
Live, during recording · Follows scroll and drawers · Placeholders too
A one-page checklist · A paste-in for AI instructions · Diagnostics after every rehearsal
It drives the real product, not screenshots of it. When your app changes, re-rehearse the chapter and export again.
Open menus, type into fields, draw on a floor plan, drag one thing onto another. Written the way you’d explain it, understood without a prompt.
Clips are generated once and kept; only a sentence you change is made again, and delivery carries across sentences.
Your app, your data, your tutorials stay local. Your account holds a plan and a count of minutes, kept in Canada, and nothing else.
Tuti Tuti is built by Pandacat in Canada and keeps its accounts in Canada. Nothing you record ever reaches a server. The privacy policy is written to PIPEDA and to Quebec’s Law 25, and public-sector buyers can pay by purchase order in Canadian dollars. For government agencies · Data residency · PIPEDA and Law 25
One sentence per beat. Name things the way they appear on screen. That’s most of the craft, and the guide has the rest.
Warehouses Now let's take a look at the warehouse map. Each aisle and shelf is laid out like your real space. Click New warehouse at the top right. For example, type "Main floor" into the Name field. Then click Create. Here you'll see the new warehouse with its first aisle. Adding stock Now let's move on to adding items. You can add items by photo, voice, or typing. Click Add item at the top right. Now choose Photo. Point the camera at the product and the details fill themselves in.