Fo MCP lets your AI control lamps, vacuums and other devices in the physical world
The Hook
“You're going to be late, don't make me remind you a third time!”
Problem Agitation
“(No spoken dialogue — ambient upbeat electronic beat)”
Product Reveal
“Fo, find me a table for 4 on Friday. Done. Piccolo Forno Friday at 9, table for four.”
Feature Teaser
“Sam needs a clean. Can you start?”
Feature Teaser
“(No spoken dialogue — upbeat synth outro)”
Call to Action
“(No spoken dialogue — audio chime)”
Synthesizable in Remotion · key components of the hardware-ai-agent-launch template.
Rounded micro-video player embedded seamlessly inside a line of text
Wrap Remotion OffthreadVideo inside rounded inline-flex spans with fixed aspect ratio and hidden overflow.
Frosted rounded bubble that expands with spring physics and displays avatar and transcript
Use spring() to drive translateY and opacity; apply backdropFilter blur and translucent white background.
Square monochrome tiles fade and snap together sequentially to form a glyph
Map an array of coordinates, staggered with frame offsets and interpolate() on opacity and scale.
Purpose. Demonstrate personality and immediate utility in a real home setting.
Execution. Robotic desk lamp vocalizes a calendar reminder.
Purpose. Communicate multi-device versatility rapidly.
Execution. Horizontal kinetic typography with embedded micro-videos.
Purpose. Show Fo executing complex agent tasks via MCP.
Execution. Voice booking on the tabletop puck with instant UI confirmation.
Purpose. Highlight cross-device autonomous control.
Execution. Voice command triggers robotic vacuum sequence.
Purpose. Cement protocol identity and parent brand.
Execution. Minimal title lockup followed by pixel-assembled brand logo.
Objective
Launch an AI-enabled desktop micro-robot for developer workflows.How the recipe adapts
Replace the lamp with a mini robot reacting to failing unit tests, keep the inline text scroller for IDE integrations, and show the puck approving Git pull requests.Scene durations (6 scenes, 25s)
Key Remotion components
Introducing Griffin, the first AI to pass the video Turing test
The most emotive and controllable TTS model
We just released QVAC TranscribePsy, a speech-to-text model small enough to run inside a phone app. It covers 25 European languages on a laptop and 8 on a phone, and on an iPhone 17 Pro Max's Ne