Learn from narrated screenshares
Uses a narrated screenshare of a task to learn the process.
Automation / product dossier
Learns back-office workflows from narrated screenshares and runs logged, permission-scoped agents.
Product brief
Caddi turns narrated screenshares into production agents that run back-office work across your real tools. Show the task once, and Caddi learns the process, builds the automation, and lets you update it in plain English. Unlike RPA or workflow builders that require scoping, developers, and step-by-step setup, Caddi combines AI reasoning with deterministic execution, with every run logged and every permission scoped.
Why we selected it
A concrete approach to operational agent deployment: demonstrate a back-office task once, then get logged, permission-scoped automation. It is especially relevant as companies focus on turning AI access into repeatable,
Product preview saved with our daily selection
Capability scan
Only capabilities supported by the product information we collected are listed here.
Uses a narrated screenshare of a task to learn the process.
Creates agents to run back-office work across an organization's tools.
Lets users update an automation through plain-English instructions.
Provides logs for each run and scoped permissions.
Best-fit use cases
FAQ
Caddi is an agent-building product that turns narrated screenshares into production agents for back-office work.
You show and narrate the task once; Caddi is described as learning the process and building the automation.
Yes. The product description says automations can be updated in plain English.
Caddi states that every run is logged and every permission is scoped.