From hot news to post-ready content.

End-to-end, in one agent.

The short-form era turned every brand and every creator into a multi-platform publisher. Content Agent is the tool I wanted. It is one system that sits on top of this whole workflow and serves companies, agencies, and solo creators with batch content production across every major platform.

Most AI content tools have one trick. They turn anything into an AI-generated video. But a few seconds of generated footage cannot explain a market move, a regulatory shift, or a tech announcement, and matching abstract news to convincing visuals is a losing game. Content Agent starts from the opposite premise. For every hot topic it fetches, an LLM judge picks the optimal format. A video only when the topic itself is visual, like a runway show or a sports moment. An image with voiceover for narrative news. A multi-image slideshow with TTS narration for finance and tech. A carousel for list-style content. A plain text post for a one-line take. Each format runs through a dedicated pipeline. Runway Aleph2 and fal.ai (Seedance, Veo3, Kling, Hailuo) handle the visuals. Edge-tts handles multilingual voiceover. ffmpeg assembles slideshows with Ken Burns motion. DeepSeek drives ideation, copy, and decisioning.

Every step stays human-in-the-loop. You can edit the scene prompt, rewrite the narration, swap image prompts in raw JSON, or override the agent's format choice before anything generates. A SQLite-backed feedback loop closes the system. After publishing, log a post's actual performance. DeepSeek distills your winners into platform-specific style guides that feed back into the next batch as few-shot context, so the system gets sharper with every post you ship.