Spec Driven Development
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Description from the maker (original English)
Curated real launch videos split into timed sections (hook, problem, feature, proof, CTA).
These three are never merged into one score: opinions, measurements and popularity are different things.
We have not written a plain-language summary for this one yet. Below are the facts we read from the registry.
Pick the AI app you use, then copy the settings.
You need
Which app do you use?
Pick your app above. We remember it, so next time the button says “Install in …”.
Works in Claude on the web, desktop and mobile:
https://scrollsnacks.com/mcpRun this in your terminal
claude mcp add --transport http scrollsnacks https://scrollsnacks.com/mcpPaste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json
{
"mcpServers": {
"scrollsnacks": {
"url": "https://scrollsnacks.com/mcp"
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
Paste into .vscode/mcp.json in your project
{
"servers": {
"scrollsnacks": {
"type": "http",
"url": "https://scrollsnacks.com/mcp"
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
ChatGPT connects only to servers on the internet:
https://scrollsnacks.com/mcpMenu names can differ slightly by plan and over time.
Your choice stays on this device in a cookie; no account needed.
These settings are generated from the maker’s registry entry. If the README says otherwise, follow the README.
The server asks you to sign in before it lists its tools.
We check information anyone can see; we have not read the code. Checks that do not apply are left out of the count. · Checked Nov 6, 2026
The registry name was verified against the domain scrollsnacks.com (the registry checks domain ownership)
We have not read this repo from GitHub yet
We have not read this repo from GitHub yet
Responded in 858 ms (HTTP 401)
Declares OAuth sign-in, so your AI app can take you to log in
Not checked yet; we read the tool list when we check the endpoint
Compared with every active server in this category in the registry. Stars count only when the publisher owns the repo.
It passed 3 of 3 applicable basic checks. We have not read the code. Before adding a key or token, see what access the server asks for and use a key with only the permissions you need.
The registry does not list a required key, though some servers ask you to sign in on first use.
It runs in the maker’s cloud. Connect over the internet with nothing to install, but what you send passes through the maker’s servers.
The GitHub repo last changed –, and the latest version (1.0.0) reached the registry yesterday.
Apps that support MCP, such as Claude Desktop, Claude Code, Cursor and VS Code. It connects by URL, so apps that accept remote MCP servers work too, such as ChatGPT in developer mode.
In the same category: Spec Driven Development, Remote Desktop Commander, LickityClick
Servers in the same category from other makers, sorted by stars.
Reading a card:★ GitHub starsiThree pillars: reviews · checks · usageiUpdated = latest version published to the registryi
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Remote Desktop Commander: use files, terminal and processes on your own computers from any AI app.
Create, edit and track LickityClick short links, campaigns and click stats from your AI assistant.
Copied from the maker’s repo as written. PopNx MCP has not reviewed or endorsed it. HTML, images and scripts are removed. Links lead off this site.
Scrollsnacks is a curated library of real launch videos from products and studios. Each video is split into timed sections (hook, problem, feature, proof, CTA), so you can study and reuse what works when you plan your own launch video.
This repo is the public home of the hosted Scrollsnacks MCP server. It holds docs, the registry server.json and a few agent skills. There is no server code here; the server runs at:
https://scrollsnacks.com/mcp
Streamable HTTP, OAuth 2.1 with dynamic client registration, scope library:read. You sign in with your Scrollsnacks account the first time your client connects.
Scrollsnacks helps you plan and script. It does not make, generate or render video. Every video in the library belongs to its brand (and studio, where we can confirm it).
Claude (web and desktop): Settings > Connectors > Add custom connector, paste https://scrollsnacks.com/mcp, then sign in.
Claude Code:
claude mcp add --transport http scrollsnacks https://scrollsnacks.com/mcp
Or install this repo as a Claude Code plugin to get the server and the skills together (see .claude-plugin/plugin.json and .mcp.json).
Cursor: add to ~/.cursor/mcp.json (or .cursor/mcp.json in a project):
{
"mcpServers": { "scrollsnacks": { "url": "https://scrollsnacks.com/mcp" } } }
VS Code (Copilot agent mode): add to .vscode/mcp.json:
{
"servers": { "scrollsnacks": { "type": "http", "url": "https://scrollsnacks.com/mcp" } } }
ChatGPT: custom remote MCP connectors work through developer mode (Settings > Apps & Connectors > Advanced settings > Developer mode), where available on your plan. Create a connector with https://scrollsnacks.com/mcp and OAuth.
Some tools and fields are only available on the Pro plan; the tool descriptions say which.
| Tool | What it does |
|---|---|
search_videos | Find launch videos by text, category, style tag, brand or studio. Start here to pick a reference. |
get_video | One video by id: metadata and timed sections. view: 'blueprint' gives beats with durations, words per second and category norms. |
search_sections | Find sections across videos (CTAs, proof, feature demos) by type, effect, technique, text, category, brand or duration. |
get_section_media | One section in detail: motion strip image, transcript, on-screen text, optional short MP4. |
search_hooks | Opening hooks grouped by formula, with category medians for hook length and pace. |
get_benchmarks | Planning norms for a category: median duration, hook length, when the product is named, CTA position, section shares. |
list_section_types | The filter vocabulary: section types, effect labels and style tags. |
create_stitch | Make a stitch: your ordered list of beats from real sections plus your own shots, with a build note per beat. |
get_stitch | Read a stitch, or get it as a production shot list or build prompt. |
update_stitch | Change your stitch: title, aspect, visibility or beats. |
add_comment | Add, edit or delete a note pinned to a beat, a moment or a spot in your stitch. |
list_stitches | Your stitches, newest first. |
delete_stitch | Permanently delete one of your stitches. |
The server also offers the prompts remake_video, find_examples and audit_script.
The skills/ folder has Agent Skills (SKILL.md with YAML frontmatter) that use the tools above. Copy a folder into your client's skills directory (for example ~/.claude/skills/ or .cursor/skills/), or install the repo as a Claude Code plugin.
| Skill | Use it to |
|---|---|
| launch-video-script | Write a timed script and shot list from 3 to 5 matching launches. |
| launch-video-teardown | Break down one launch section by section. |
| hook-finder | Find strong hooks for a category and draft yours. |
| storyboard-from-formats | Turn a chosen format into a storyboard with a stitch. |
| cta-check | Check that your script ends with one clear ask. |
Skills and docs in this repo are MIT licensed. Launch videos belong to their brands and studios.