About ClipmivoAI
ClipmivoAI provides API, CLI, local MCP server, and agent Skill tools for creating AI videos. It supports live model discovery, credit quotes, task submission, and media handling for developers and agents using a ClipmivoAI account.
Key features
- 01Live video model discovery
- 02Credit balance and generation quotes
- 03Video task submission and status checks
- 04Download generated media
- 05Local MCP server and agent Skill
Use cases
- 01Compare available models before generation
- 02Check spending before submitting a video task
- 03Generate and retrieve videos from an application
- 04Connect an MCP client to ClipmivoAI
MCP support
Provides a local stdio MCP server that connects to the ClipmivoAI API using a scoped API key. MCP clients can use it alongside the CLI and agent Skill.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEClipmivoAI developer tools
Create AI videos through the ClipmivoAI API, CLI, local MCP server or agent Skill. All clients use https://clipmivoai.com/api/v1, your scoped API key and your ClipmivoAI account balance.
English · 简体中文 · Website · API guide · OpenAPI
Choose your interface
| Interface | Start here |
|---|---|
| REST API | Authentication, quotes and task submission |
| CLI | Install and use the CLI |
| MCP | Connect a local stdio server |
| Agent Skill | clipmivo-video |
Sign in at ClipmivoAI, create a scoped key in the API / CLI (Developer) workspace and add credits through the website. Read operations require video:read, quotes and generation require video:write, and media uploads require files:write. API requests create video tasks; they do not purchase credit top-ups.
Install
Requires Node.js 22+ and npm. Published downloads are hosted in GitHub Releases; these commands do not depend on an npm registry listing:
npm install -g https://github.com/BarneyD66/clipmivo-tools/releases/download/v0.1.1/clipmivo-cli-0.2.7.tgz
npm install -g https://github.com/BarneyD66/clipmivo-tools/releases/download/v0.1.1/clipmivo-mcp-0.1.8.tgz
clipmivo --version
clipmivo-mcp --version
Set CLIPMIVO_API_KEY using your shell or secret manager. Never put a real key into this repository. For development:
git clone https://github.com/BarneyD66/clipmivo-tools.git
cd clipmivo-tools
npm ci
npm test
node cli/bin/clipmivo.mjs models --json
First video
clipmivo models --json
clipmivo balance --json
clipmivo quote --file examples/request.json --json
Review the current quote and confirm your spending limit. Follow the CLI quote confirmation example to add quoted_credits to a separate request file, then submit:
clipmivo generate --file quoted-request.json --idempotency-key my-first-shot-001 --wait --timeout 600 --json
clipmivo get TASK_ID --json
clipmivo download TASK_ID --output result.mp4 --json
Keep the same confirmed request and idempotency key after an uncertain submission. A wait timeout is not a failed video; resume status queries instead of creating another task. Model options and prices can change: use live model discovery and quotes.
Credit billing and your application
Show credits_to_hold directly as Credits; do not multiply an API credit quote by 100. Credits are reserved before generation, then settled using actual usage where applicable. Unused reserved credits return to the balance. The reference conversion is 100 Credits = US$1; top-ups are purchased separately on the website.
For a customer-facing app, keep the ClipmivoAI key on your server. Keys on one account share its balance and account-level resources; they are not isolated reseller subaccounts. See application integration.
MCP
Configure a client that supports local stdio MCP:
{
"mcpServers": {
"clipmivo": {
"command": "clipmivo-mcp",
"env": {
"CLIPMIVO_API_KEY": "YOUR_SCOPED_API_KEY",
"CLIPMIVO_FILES_DIR": "/absolute/path/to/video-files"
}
}
}
}
Store the real key in your client's secret settings. On Windows use an existing directory such as C:/Users/you/Videos/Clipmivo; see the MCP guide if your client cannot resolve command shims. This is a local server, not a hosted MCP OAuth endpoint.
Skill
Copy skills/clipmivo-video into the skill directory supported by your agent. For Codex, use ~/.codex/skills/clipmivo-video. Restart or reload skills, then ask:
Use $clipmivo-video to create a video of a paper boat on a pond. Maximum generation spend: 25 Credits.
The Skill can use the installed MCP or CLI. It preserves existing budget authorization and requires a valid quote before paid submission. Installing the Skill does not create an API key or grant credits.
Support
Use GitHub issues for reproducible client bugs without credentials or private media. Account, payment and billing questions: [email protected]. Service use remains subject to the website's terms and privacy policy.
This repository contains public client tools and documentation, not the website backend.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.