Who is agent-media for?

Suits creators and developers who already use an AI agent such as Claude Code or Cursor and are happy to pay credits to a hosted service; knowing basic npm and API tokens helps.

What can you build with agent-media?

Ask Claude Code to make a talking-person product teaser from a short script. Turn a long monologue into a multi-take vertical video for TikTok or Reels. Produce a two-person podcast clip from a scripted dialogue between saved characters. Reuse a saved character across several videos to keep the same face and voice. Generate a video and schedule it to a connected social channel from the terminal.

gitroomhq /

agent-media

Skill and MCP server that lets AI agents make lip-synced vertical UGC videos

82 stars
3 stars since 3 Oct
Updated 17 days ago
Beginner friendly
Image, video & audio

Open on GitHub (opens in a new tab) Website (opens in a new tab)

agent-media is a plugin for AI agents that makes short vertical videos in the style of user-generated content (UGC). You give it a script and say who should speak: a text description, a photo, or a saved character. It returns a finished MP4 with the person speaking your lines, lip-synced.

It works through a hosted API. Your agent calls one tool, make_ugc, which starts a job. You poll the job until it succeeds and read the video URL from the result. Short scripts become one clip. Long scripts are split into several takes and joined, and the README says they are never trimmed. A b-roll URL gives a narrated overlay. Captions are off unless you ask for them.

You can reach it as a Claude Code plugin, through the skills installer, as an MCP server, over plain REST, or with a CLI. A second skill, make_podcast, puts two saved characters in one room and cuts between them as they talk. The repo is a mirror of a private monorepo, so it holds skills and docs, not the generation code. Each run costs credits bought on the agent-media site.

What it does

One tool: make_ugc
Takes a script plus a person description, photo, or saved character and returns a finished vertical video. With no identity given, it uses a default person.
Long scripts and b-roll
Long monologues are rendered as several takes and stitched together instead of being trimmed. Passing a broll_url gives a narrated overlay.
Podcast episodes
make_podcast takes two saved characters and a list of A/B dialogue turns. It keeps each speaker's seat, look and voice consistent across cuts and outputs a 9:16 video.
Many ways to connect
Install through npx skills, a Claude Code plugin, an MCP server, REST calls, or the agent-media CLI. One Bearer token covers them all.
Publish to social
Connect TikTok, Instagram or X by OAuth, list channels, and post or schedule a generated video through REST, the CLI, or MCP tools.

What you could build

Ask Claude Code to make a talking-person product teaser from a short script.
Turn a long monologue into a multi-take vertical video for TikTok or Reels.
Produce a two-person podcast clip from a scripted dialogue between saved characters.
Reuse a saved character across several videos to keep the same face and voice.
Generate a video and schedule it to a connected social channel from the terminal.

Get started

Install the skills into your agent, then log in with the CLI or copy the ma_ token from the dashboard. Buy credits at agent-media.ai, then ask your agent to make a UGC video from a script.

npx skills add gitroomhq/agent-media

npm i -g agent-media-cli && agent-media login

Commands are copied from the project's README. Read it on GitHub for the full setup.