Katto

Katto is an AI-powered video clipping tool built in Japan that converts long-form videos into post-ready vertical short clips. It detects the best moments, reframes to 9:16, adds word-synced captions in 99 languages, scores each clip for viral potential, and publishes directly to 7 platforms — all from a single upload or YouTube link. Katto also exposes a REST API and a hosted MCP server, making it accessible to AI agents like Claude and Cursor for fully programmatic clip generation.

What It Is

Katto sits in the AI video repurposing category, targeting podcasters, sports creators, interviewers, gamers, and anyone who needs to turn long recordings into short-form social content at scale. The core workflow is: paste a link or upload a file, let the AI scan for the best moments, review ranked clips in the editor, and export or schedule to platforms like YouTube, TikTok, Instagram, LinkedIn, Threads, Bluesky, and Mastodon. The tool positions itself against OpusClip and Klap with flat pricing (no credits), transparent scoring, and a reframe engine that handles sports, gaming, and multi-speaker content — not just talking heads.

How the Scoring and Editing Work

Every clip Katto generates is scored on four axes — hook, flow, value, and trend — and ranked by viral potential so the best moments surface automatically. The editor exposes all AI decisions (reframe targets, captions, hooks, B-roll, emojis) and lets users override any of them. Text-based editing lets users delete or reorder transcript sentences and the video cuts follow, similar to editing a document. A prompt-first mode lets users direct the AI by intent, topic, hook, or audience angle rather than relying on random moment detection.

Miru AI Reframe and Caption Engine

Katto's proprietary reframe engine, called Miru AI, follows speakers, faces, and action shot-by-shot, supporting multi-face layouts, speaker splits, and shot changes. It outputs 9:16, 4:5, and 1:1 formats. The caption engine is word-synced with karaoke-style highlighting, supports 99 languages with AI proper-noun correction, and offers up to 40 style presets on paid plans. AI dubbing translates clips into 19 languages with a natural voice.

REST API and MCP Server for Developers and Agents

Katto ships a REST API (OpenAPI 3.1 spec), webhooks, and a hosted MCP server accessible via a single npx command. This lets developers wire Katto into their own stacks or hand clip generation to AI agents — the site specifically calls out Claude and Cursor as compatible agents. The API and MCP access are included in the Creator plan. This developer surface distinguishes Katto from most consumer-only video clipping tools.

Platform and Export Capabilities

• Supported input: YouTube links, MP4, MOV, WebM, MKV up to 90 minutes
• Upload limits: 1 GB on Free, 5 GB on Creator
• Output: 720p on Free, 1080p on Creator (4K planned for the upcoming Studio tier)
• Direct scheduling and publishing to 7 platforms
• SRT subtitle export
• Audio features: silence removal, vocal cleanup, background music
• Brand templates, folders, and favorites for organization

Current Status

Katto is live and actively accepting signups with a free tier requiring no credit card. The site advertises a founding-price promotion for the first 50 Creator seats. A third tier called Studio — targeting power users and small teams with 4K export, 50 videos/month, and 3 team seats — is listed as "coming later this year." The changelog and roadmap pages are live, indicating active product development.