Introduction
Decodx takes one screen recording and produces two things: a finished video — cut, narrated, captioned, with chapters and a poster frame — and a written step-by-step guide that matches it. Both come from the same source, so they never drift apart.
Who it's for
Teams that ship the same kind of video over and over — product tutorials, onboarding walkthroughs, release notes, support answers, internal enablement. The kind of video that is worth making and then immediately goes out of date.
What makes it different
The video and the guide come from one recording. Most tools give you one or the other. Writing the guide by hand afterwards is where teams give up.
Re-rendering is cheap. Every expensive piece — the capture, the narration audio, the avatar clip — is produced once and reused. Change one line of script and only that scene is re-made. A correction costs seconds, not another afternoon.
An agent can drive the whole thing. Decodx exposes an MCP server and a REST API, so Claude, your IDE, or your own service can create projects, set narration and render — without a person in the loop.
What a render gives you
| Artifact | Notes |
|---|---|
final.mp4 | The finished video |
| Captions | Burned in, and as a sidecar file |
| Chapters | Derived from your scenes |
| Poster frame | For thumbnails and social cards |
| A written guide | Step-by-step, exportable, embeddable |
| A share link | Public or restricted, with view analytics |