mcpField
I
imaginevid-ai-generation

imaginevid-ai-generation

Remote
imaginevid.ioMedia & Design0Updated 12w ago

About imaginevid-ai-generation

imaginevid-ai-generation distributes open-source packages for the ImagineVid image, video, and music generation platform, including a REST client, CLI, and local MCP adapter. It supports hosted OAuth or a local API key, with generation quotes and explicit credit confirmation.

Key features

  1. 01List models and check credits
  2. 02Quote and submit generation requests
  3. 03Retrieve generation status
  4. 04Upload assets and reference owned asset IDs
  5. 05Require credit confirmation and idempotency keys

Use cases

  1. 01Review available models and account credits
  2. 02Check cost and credit usage before generation
  3. 03Submit generation requests and track their status
  4. 04Upload assets and reference them in generation requests

MCP support

It provides an OAuth-protected hosted Streamable HTTP MCP server and a local stdio adapter using an API key. The tools cover models, credits, quotes, and generation requests.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://imaginevid.io/api/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Packages
npmimaginevid-mcp0.1.0

Tools

Details

From the project README

ImagineVid Agent Tools AI

Open-source distribution packages for ImagineVid, an AI generation platform for images, videos, and music. This repository contains the typed REST client, a CLI, and a local stdio MCP adapter. The hosted MCP server remains the source of truth for identity, OAuth consent, scopes, credits, quotes, idempotency, owned assets, and provider lifecycle.

Choose a connection

Hosted MCP

Use the OAuth-protected Streamable HTTP endpoint when your host supports remote MCP servers:

{
  "mcpServers": {
    "imaginevid": {
      "url": "https://imaginevid.io/api/mcp"
    }
  }
}

The host handles OAuth and explicit consent. The endpoint exposes models_list, credits_get, generation_quote, generation_create, and generation_get. See the ImagineVid website for the product and account flow.

Local stdio MCP

After the npm package is released, configure a local host with:

{
  "mcpServers": {
    "imaginevid": {
      "command": "npx",
      "args": ["-y", "imaginevid-mcp"],
      "env": {
        "IMAGINEVID_API_KEY": "${IMAGINEVID_API_KEY}",
        "IMAGINEVID_BASE_URL": "https://imaginevid.io"
      }
    }
  }
}

The local adapter uses a scoped API key, never prints it, and never retries an ambiguous submission_unknown result. It accepts product capability values and owned asset IDs, not provider parameters, arbitrary remote URLs, or inline base64 media.

CLI

After the npm package is released:

npm install --global imaginevid-cli
export IMAGINEVID_API_KEY="..."
imaginevid models list
imaginevid credits

The CLI also supports assets upload, generate, generations get, and generations wait. generate quotes first and requires an exact explicit credit confirmation plus a stable idempotency key.

Development

pnpm install
pnpm test
pnpm build

The public package metadata and release files live in:

For product documentation, generation tools, and account access, visit the ImagineVid website. Source issues belong in this GitHub repository.

Security and license

Do not commit API keys or account configuration. The hosted endpoint uses OAuth; the local adapters read credentials from the environment or an owner-only local config file. Report package or security issues through the ImagineVid support page.

The public distribution code is available under MIT-0.

Comments

Markdown supported

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.