Log inSubmit server
Kie MCP logo

Kie MCP

Connects MCP clients to kie.ai for image, video, audio generation, model discovery, file handling, and task management.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install Kie MCP

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "kie-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "kie-mcp"
      ],
      "env": {
        "KIE_API_KEY": "YOUR_VALUE_HERE"
      }
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

Values are written into the snippet above so you can paste it as is. They never leave your browser.

Automated check passed— started and listed 47 tools correctly (2d ago).

Overview

The kie-mcp MCP server connects MCP clients to the kie.ai generation API, exposing more than 60 image models, 85 video models, and 20 audio tools. It provides model search, research-backed recommendations, pricing information, generation workflows, file uploads, task status, credit checks, and asset listing. Run it locally over stdio or remotely over authenticated Streamable HTTP. Reach for it when an agent needs to select and use kie.ai multimedia models from a conversational workflow.

Use cases

  • Generate images, videos, music, or speech through kie.ai
  • Search multimedia models by capability or cost
  • Route creative work with vertical profiles
  • Upload reference media for generation workflows
  • Check task status, credits, and generated assets

Key features

  • Image, video, and audio model access
  • Research-backed model discovery and routing
  • Pricing and credit-aware selection
  • Vertical profiles and MCP prompts
  • File upload and asset management
  • Stdio and authenticated HTTP transports

Tools47

What Kie MCP lets your agent do. Expand a tool for its parameters.

Verified live~15.8k tokens of context

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked 2d ago

Recent health check succeeded.

6/6 checks healthy over the last 40d

What it can do

From the 47 tools this server returned when we ran it.

Can delete or modify data · 1Writes files · 4Makes network requests · 8Read-only tools · 42
Network at startup3 connections
Files written at startupNone
Install scriptsNone
npm provenanceSigned build
Package age161 days
Maintainers1

Observed 2026-10-20 in an isolated sandbox. A description of reach, not a security audit.

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 3d ago via OSV.dev · kie-mcp (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars7Stargazers on the source repository.
Last commit3d agoMost recent push to the default branch.
Availability100%Our rolling endpoint + install checks that succeeded.
Install checkPassedOur sandbox started it and listed its tools.
Tools exposed47Callable tools this server registers over MCP.
Directory activity2 viewsConfig copies, upvotes, and views on AllMCPs.

How Kie MCP works

What kie-mcp MCP server does

The kie-mcp MCP server connects an MCP client to kie.ai’s generation API. It covers image, video, and audio workflows, with model catalogs spanning more than 60 image models, 85 video models, and 20 audio tools. Supported model families include image generators and editors, video generation and transformation systems, avatar and lip-sync tools, music generation, text-to-speech, dialogue, audio isolation, and speech-to-text.

The server also exposes utility operations for uploading files, checking task progress, inspecting credits, and listing raw assets. Generated files can be written to a configured project location or to an absolute directory supplied for a file-writing call.

How it works

The kie-mcp MCP server gives the client model metadata that includes capabilities, pricing in credits and USD, research notes, prompt guidance, limitations, and competitor comparisons. Its model search can match capability tags, descriptions, and research fields, so queries such as a need for lip sync, architecture, or a low-cost video model can narrow the catalog.

Vertical profiles provide domain-specific intake questions, model routing, deliverable costs, prompt formulas, and multi-tool workflows. The documented verticals include architecture and interiors, video game assets, advertising, product imagery, film, e-commerce, brand design, editorial work, social video, and audio branding. Profiles are also available as MCP prompts for supported clients.

Use stdio for a local process, including Claude Code or Claude Desktop. HTTP mode exposes an MCP endpoint for remote clients and requires a bearer token. HTTP listens on loopback by default; remote access requires a tunnel or reverse proxy and appropriate host configuration.

Setup and configuration

Install Node.js 18 or newer, obtain a kie.ai API key, clone the repository, and run npm install. Start the local server with Node and provide KIE_API_KEY in the MCP client configuration. KIE_PROJECT_ROOT optionally sets the default output location, which otherwise uses the server’s working directory.

HTTP deployments require KIE_MCP_AUTH_TOKEN; the server refuses to start in HTTP mode without it. KIE_MCP_PORT changes the default port of 3100, while KIE_MCP_HOST, KIE_MCP_ALLOWED_HOSTS, and KIE_MCP_ALLOWED_ORIGINS control binding and request restrictions. KIE_MAX_CREDITS_PER_CALL can impose a default spending limit before a request is submitted. Additional variables control Suno routing, ElevenLabs availability, and Suno callbacks.

Tools and capabilities

  • Search models by capability, description, or research information.
  • Inspect model costs, prompt techniques, weaknesses, and comparative notes.
  • Generate and edit images, videos, music, speech, dialogue, and other audio assets.
  • Use video extension, 1080p or 4K upscaling, avatar, and lip-sync workflows where supported by the selected model.
  • Upload media from a URL or base64 input.
  • Check task status, credits, and raw generated assets.
  • Apply vertical profiles to structure domain-specific generation work.

Limitations and notes

Availability and behavior depend on the corresponding kie.ai models and endpoints. ElevenLabs tools are paused by default unless explicitly re-enabled. Suno can use either the newer market API or legacy endpoints, with some functions remaining on legacy endpoints.

Treat the HTTP bearer token as a secret: someone holding it can spend kie.ai credits and upload media from the server’s disk. Keep HTTP bound to loopback unless it is protected by a suitable firewall, tunnel, or reverse proxy, and configure allowed hosts and origins for the deployment.

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Kie MCP

kie-mcp is an MCP server that connects Claude and other MCP clients to the kie.ai generation API. Its main tools cover image, video, and audio model discovery and generation, plus file upload, task status, credit checks, raw asset listing, model profiles, and cost-aware recommendations.

Install Node.js 18 or newer, clone the repository, run npm install, and configure KIE_API_KEY in your MCP client. Start server.mjs locally over stdio, or use --http with an authentication token for HTTP mode.

Yes, kie-mcp supports Claude Desktop through its stdio transport. Add the server command and server.mjs path to the client’s MCP configuration, then provide the required KIE_API_KEY environment variable.

Run server.mjs with --http and provide KIE_MCP_AUTH_TOKEN; the default port is 3100 and the default bind address is 127.0.0.1. Configure the client with the /mcp URL and a Bearer authorization header, using a tunnel or reverse proxy for remote access.

Yes, kie-mcp exposes model pricing and can enforce a default credit limit with KIE_MAX_CREDITS_PER_CALL. A per-call max_credits value can override that default, and model search can identify cheaper options for a use case.

Related MCP servers

All in Multimedia Process All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewKie MCP AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/kie-mcp?style=directory)](https://allmcps.com/mcp/kie-mcp)
HTML Embed
<a href="https://allmcps.com/mcp/kie-mcp"><img src="https://allmcps.com/api/badge/kie-mcp?style=directory" alt="Kie MCP on AllMCPs" /></a>