
Kie.ai MCP Server
Enables AI-powered image generation and editing with Nano Banana, plus professional video creation with Veo3, all through Kie.ai's APIs. Includes task tracking, status monitoring, and 1080p video upgrades.
README
Kie.ai MCP Server
An MCP (Model Context Protocol) server that provides access to Kie.ai's AI APIs including Nano Banana image generation/editing and Veo3 video generation.
Features
- Nano Banana Image Generation: Text-to-image generation using Google's Gemini 2.5 Flash Image Preview
- Nano Banana Image Editing: Natural language image editing with up to 5 input images
- Veo3 Video Generation: Professional-quality video generation with text-to-video and image-to-video capabilities
- 1080p Video Upgrade: Get high-definition versions of Veo3 videos
- Task Management: SQLite-based task tracking with status polling
- Smart Endpoint Routing: Automatic detection of task types for status checking
- Error Handling: Comprehensive error handling and validation
Prerequisites
- Node.js 18+
- Kie.ai API key from https://kie.ai/api-key
Installation
From NPM
npm install -g @andrewlwn77/kie-ai-mcp-server
From Source
# Clone the repository
git clone https://github.com/andrewlwn77/kie-ai-mcp-server.git
cd kie-ai-mcp-server
# Install dependencies
npm install
# Build the project
npm run build
Configuration
Environment Variables
# Required
export KIE_AI_API_KEY="your-api-key-here"
# Optional
export KIE_AI_BASE_URL="https://api.kie.ai/api/v1" # Default
export KIE_AI_TIMEOUT="60000" # Default: 60 seconds
export KIE_AI_DB_PATH="./tasks.db" # Default: ./tasks.db
MCP Configuration
Add to your Claude Desktop or MCP client configuration:
{
"kie-ai-mcp-server": {
"command": "node",
"args": ["/path/to/kie-ai-mcp-server/dist/index.js"],
"env": {
"KIE_AI_API_KEY": "your-api-key-here"
}
}
}
Or if installed globally:
{
"kie-ai-mcp-server": {
"command": "npx",
"args": ["-y", "@andrewlwn77/kie-ai-mcp-server"],
"env": {
"KIE_AI_API_KEY": "your-api-key-here"
}
}
}
Available Tools
1. generate_nano_banana
Generate images using Nano Banana.
Parameters:
prompt
(string, required): Text description of the image to generate
Example:
{
"prompt": "A surreal painting of a giant banana floating in space"
}
2. edit_nano_banana
Edit images using natural language prompts.
Parameters:
prompt
(string, required): Description of edits to makeimage_urls
(array, required): URLs of images to edit (max 5)
Example:
{
"prompt": "Add a rainbow arching over the mountains",
"image_urls": ["https://example.com/image.jpg"]
}
3. generate_veo3_video
Generate videos using Veo3.
Parameters:
prompt
(string, required): Video descriptionimageUrls
(array, optional): Image for image-to-video (max 1)model
(enum, optional): "veo3" or "veo3_fast" (default: "veo3")aspectRatio
(enum, optional): "16:9" or "9:16" (default: "16:9")seeds
(integer, optional): Random seed 10000-99999watermark
(string, optional): Watermark textenableFallback
(boolean, optional): Enable fallback mechanism
Example:
{
"prompt": "A dog playing in a park",
"model": "veo3",
"aspectRatio": "16:9",
"seeds": 12345
}
4. get_task_status
Check the status of a generation task.
Parameters:
task_id
(string, required): Task ID to check
5. list_tasks
List recent tasks with their status.
Parameters:
limit
(integer, optional): Max tasks to return (default: 20, max: 100)status
(string, optional): Filter by status ("pending", "processing", "completed", "failed")
6. get_veo3_1080p_video
Get 1080P high-definition version of a Veo3 video.
Parameters:
task_id
(string, required): Veo3 task ID to get 1080p video forindex
(integer, optional): Video index (for multiple video results)
Note: Not available for videos generated with fallback mode.
API Endpoints
The server interfaces with these Kie.ai API endpoints:
- Veo3 Video Generation:
POST /api/v1/veo/generate
✅ VALIDATED - Veo3 Video Status:
GET /api/v1/veo/record-info
✅ VALIDATED - Veo3 1080p Upgrade:
GET /api/v1/veo/get-1080p-video
✅ VALIDATED - Nano Banana Generation:
POST /api/v1/playground/createTask
✅ VALIDATED - Nano Banana Status:
GET /api/v1/playground/recordInfo
✅ VALIDATED
All endpoints have been tested and validated with live API responses.
Database Schema
The server uses SQLite to track tasks:
CREATE TABLE tasks (
id INTEGER PRIMARY KEY AUTOINCREMENT,
task_id TEXT UNIQUE NOT NULL,
api_type TEXT NOT NULL, -- 'nano-banana', 'nano-banana-edit', 'veo3'
status TEXT DEFAULT 'pending',
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
result_url TEXT,
error_message TEXT
);
Usage Examples
Basic Image Generation
# Generate an image
curl -X POST http://localhost:3000/tools/call \
-H "Content-Type: application/json" \
-d '{
"name": "generate_nano_banana",
"arguments": {
"prompt": "A cat wearing a space helmet"
}
}'
Video Generation with Options
# Generate a video
curl -X POST http://localhost:3000/tools/call \
-H "Content-Type: application/json" \
-d '{
"name": "generate_veo3_video",
"arguments": {
"prompt": "A peaceful garden with blooming flowers",
"aspectRatio": "16:9",
"model": "veo3_fast"
}
}'
Error Handling
The server handles these HTTP error codes from Kie.ai:
- 200: Success
- 400: Content policy violation / English prompts only
- 401: Unauthorized (invalid API key)
- 402: Insufficient credits
- 404: Resource not found
- 422: Validation error / record is null
- 429: Rate limited
- 451: Image access limits
- 455: Service maintenance
- 500: Server error / timeout
- 501: Generation failed
Development
# Run tests
npm test
# Development mode with auto-reload
npm run dev
# Type checking
npx tsc --noEmit
# Build for production
npm run build
Pricing
Based on Kie.ai documentation:
- Nano Banana: $0.020 per image (4 credits)
- Veo3 Quality: Higher cost tier
- Veo3 Fast: ~20% of Quality model pricing
See https://kie.ai/billing for detailed pricing.
Production Tips
- Database Location: Set
KIE_AI_DB_PATH
to a persistent location - API Key Security: Never commit API keys to version control
- Rate Limiting: Implement client-side rate limiting for high-volume usage
- Monitoring: Monitor task status and handle failed generations appropriately
- Storage: Consider automatic cleanup of old task records
Troubleshooting
Common Issues
"Unauthorized" errors
- Verify
KIE_AI_API_KEY
is set correctly - Check API key is valid at https://kie.ai/api-key
"Task not found" errors
- Tasks may expire after 14 days
- Check task ID format matches expected pattern
Generation failures
- Check content policy compliance
- Verify prompt is in English
- Ensure sufficient API credits
Support
For issues related to:
- MCP Server: Open an issue at https://github.com/andrewlwn77/kie-ai-mcp-server/issues
- Kie.ai API: Contact support@kie.ai or check https://docs.kie.ai/
- API Keys: Visit https://kie.ai/api-key
License
MIT License - see LICENSE file for details.
Contributing
- Fork the repository
- Create a feature branch
- Make your changes
- Add tests if applicable
- Submit a pull request
Changelog
v1.0.0
- Initial release
- Nano Banana image generation and editing
- Veo3 video generation
- 1080p video upgrade support
- SQLite task tracking
- Smart endpoint routing
- Comprehensive error handling
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.