AI Services Market › Use cases › Content & Creative
Content & CreativeDone with OpenClaw Oct 22, 2026
Produce short WeChat videos with OpenClaw agents
A creator uses OpenClaw agents with Remotion and a cloned voice to suggest topics, write narration, generate TTS and render videos, confirming in Telegram in about 15 minutes.
- What They Did
- A cron job has one agent push five topic ideas to Telegram each morning, and the creator picks one. A writing agent then drafts a roughly 200-character narration and generates cloned-voice TTS through MiniMax on fal.ai. It extracts timestamps with local Whisper, fills a scenes file, and renders with Remotion. The creator reviews the video and confirms, and another agent publishes it.
- How It Went
- Each video took about 15 minutes, with roughly 10 of those unattended, and cost about ¥0.1. One video reached 1,595 views, and three days totaled 9,018. Before this, views averaged under 200.
- Worth Knowing
- Don't downgrade the writing agent's model to save money: switching from Opus to Sonnet made scripts much worse, and the author switched back within 6 minutes. Whisper timestamps can drift, so watch the render.
Try it yourself with OpenClaw
Act as my short-video producer for [platform, e.g. WeChat Channels]. Suggest 5 topics in my niche of [niche], and once I pick one, write a [length, e.g. 60-second] narration script in my voice. Generate the voiceover with [my TTS or cloned voice setup] and render a video with Remotion using [visual style]. Send me the finished file in Telegram and do not publish anything until I confirm.