What it does
The server gives AI agents seven tools for planning video ads before any footage is generated: build_ad_brief, generate_hooks, write_ugc_script, generate_storyboard, build_variation_plan, review_ad_script and export_adsturbo_prompt. Outputs include ad briefs, short-form hooks, mobile-first UGC scripts with shot notes, storyboard objects with scene timing, testable variation plans with risk notes, and prompts that can be pasted into AdsTurbo. It runs locally, makes no calls to the AdsTurbo API and uses no credits.
It suits teams that want structured creative planning ahead of paid video generation, such as ecommerce sellers or performance agencies drafting UGC scripts or storyboards for TikTok, Reels, Shorts, Meta or YouTube. Once a plan is approved, production work such as video generation, preview, export and iteration continues on the AdsTurbo website.
In an MCP client such as Claude Desktop, Cursor or Codex, the server runs over stdio after it is registered. The slash-style commands are prompt conventions that the agent maps to these tools. A terminal CLI offers the same workflows, reading product details from JSON files, inline JSON or stdin.
Limitations and requirements
The server is planning only: it does not generate video, access ad accounts or call AdsTurbo services. It must be registered in the MCP client, and after changing the MCP config the client must be restarted or a new session started before the tools appear. The local CLI requires the project to be built first.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
- Runs code or changes an assistant's behaviour
Sources
- Official MCP Registry first seen 13 Nov 2026 · last seen 13 Nov 2026 · its page on the source
AdsTurbo Creative MCP
Local-only MCP server and CLI for AI video ad briefs, UGC scripts, storyboards, and prompts.
v0.1.5As shown on registry.modelcontextprotocol.io when we last read it