Claude Code can make videos: it records the app, narrates with ElevenLabs, and syncs audio to video automatically
I'm a solo builder. I needed a 2-minute product demo for ClinTrialFinder — a free tool I built that matches cancer patients to clinical trials. I can fumble through OBS and iMovie, but I'm not proficient — and Claude Code does it faster. So I asked Claude Code — an agentic coding tool — to make it. And it did: a narrated walkthrough where the voiceover lands exactly on the on-screen action. I…
A solo developer required a brief promotional video for a free tool he created, dubbed ClinTrialFinder, which links cancer patients with clinical trials. Claude Code, an autonomous coding software, created the video swiftly, thereby saving the developer considerable time and effort. The developer did not manually record a screen or dub over narration. Instead, Claude Code performed these tasks. Here's how it went about it:
Firstly, Claude Code scripted a Playwright script that engaged the actual web application. This script carried out 10 steps of the patient enrollment process, recorded the final results page, and produced a deterministic, repeatable video output. Should the product undergo alterations, Claude Code would rerun the script and generate a new clip. Additionally, to mimic a real mouse pointer, a synthetic cursor was inserted.
Secondly, Claude Code composed the narration without the need for a microphone. The developer provided a script, and Claude Code utilized the ElevenLabs text-to-speech API to generate the narration in a consistent voice. If the developer wished to modify any line, Claude Code would simply edit the text and regenerate the clip, all within seconds.
Lastly, Claude Code synchronized the audio with the video, a task traditionally handled by a human editor. While the script was executing, Claude Code documented the timestamps of significant on-screen events, such as the submission click, the emergence of results, the opening of a trial, and the display of the "copy to AI" dropdown. These timestamps were then used to align the narration phrases with their corresponding moments in the video. This process was achieved automatically, making the synchronization seamless.
The entire process is streamlined into a single code conversation with the agent, rendering it repeatable, deterministic, and faster than manual methods. Not only does this example reflect the development of ClinTrialFinder, but it can be adapted to produce narrated, synced demos for any web application, provided there is an agent available.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.
