{"_id":"@contextarchitect/kie-ai-mcp-server","name":"@contextarchitect/kie-ai-mcp-server","dist-tags":{"latest":"3.2.0"},"versions":{"3.2.0":{"name":"@contextarchitect/kie-ai-mcp-server","version":"3.2.0","description":"MCP server for Kie.ai APIs — fork of @felores/kie-ai-mcp-server with OpenAI GPT Image 2 support added (25 AI tools: Nano Banana image generation/editing, GPT Image 2 text-to-image/image-to-image, Veo3 video generation, Suno music generation, ElevenLabs TT","main":"dist/index.js","bin":{"kie-ai-mcp-server":"dist/index.js"},"type":"module","scripts":{"build":"tsc","start":"node --env-file=.env dist/index.js","dev":"tsx --env-file=.env src/index.ts","test":"jest","prepublishOnly":"npm run build"},"keywords":["mcp","kie.ai","ai","image-generation","video-generation","music-generation","text-to-speech","nano-banana","gpt-image-2","veo3","suno","elevenlabs","bytedance","qwen","runway","midjourney","wan","hailuo","openai-4o-image","flux-kontext","flux-2","recraft","ideogram","topaz","model-context-protocol","ai-api","unified-ai"],"author":{"name":"contextarchitect","url":"fork of work by felores"},"license":"MIT","repository":{"type":"git","url":"git+https://github.com/contextarchitect/kie-ai-mcp-server.git"},"homepage":"https://github.com/contextarchitect/kie-ai-mcp-server#readme","bugs":{"url":"https://github.com/contextarchitect/kie-ai-mcp-server/issues"},"publishConfig":{"registry":"https://registry.npmjs.org/"},"dependencies":{"@modelcontextprotocol/sdk":"^0.4.0","sqlite3":"^5.1.6","zod":"^3.22.4"},"devDependencies":{"@types/jest":"^29.5.8","@types/node":"^20.10.0","@types/sqlite3":"^3.1.11","jest":"^29.7.0","ts-jest":"^29.4.6","tsx":"^4.6.0","typescript":"^5.3.0"},"engines":{"node":">=18.0.0"},"_id":"@contextarchitect/kie-ai-mcp-server@3.2.0","gitHead":"e3298928921fa2faf6d8929659a94daae6daad39","types":"./dist/index.d.ts","_nodeVersion":"22.22.0","_npmVersion":"10.9.4","dist":{"integrity":"sha512-iki9mREvrjIcuguAV7vx2C6zCQ4Iy4kmk140uMcmqxhnVi6XuCHcPE40cHNoMI3ReQRPxMOHjE9Ug6v1xrX+XQ==","shasum":"9bf6a509ef19f971d8678d6ab7e900caad7ff3f6","tarball":"https://registry.npmjs.org/@contextarchitect/kie-ai-mcp-server/-/kie-ai-mcp-server-3.2.0.tgz","fileCount":13,"unpackedSize":486527,"signatures":[{"keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U","sig":"MEYCIQDQW57J14mbaQ7gwudJdNcpDwuSxvq1WrXzfGtvQAyLWAIhAOde3KyHO4j/d+JNsd11QZl8UsjqHOkB6Z860LuVZj0e"}]},"_npmUser":{"name":"contextarchitect","email":"hilal@contextarchitect.ai"},"directories":{},"maintainers":[{"name":"contextarchitect","email":"hilal@contextarchitect.ai"}],"_npmOperationalInternal":{"host":"s3://npm-registry-packages-npm-production","tmp":"tmp/kie-ai-mcp-server_3.2.0_1776966352416_0.8727513828054654"},"_hasShrinkwrap":false}},"time":{"created":"2026-04-23T17:45:52.258Z","3.2.0":"2026-04-23T17:45:52.613Z","modified":"2026-04-23T17:45:52.879Z"},"maintainers":[{"name":"contextarchitect","email":"hilal@contextarchitect.ai"}],"description":"MCP server for Kie.ai APIs — fork of @felores/kie-ai-mcp-server with OpenAI GPT Image 2 support added (25 AI tools: Nano Banana image generation/editing, GPT Image 2 text-to-image/image-to-image, Veo3 video generation, Suno music generation, ElevenLabs TT","homepage":"https://github.com/contextarchitect/kie-ai-mcp-server#readme","keywords":["mcp","kie.ai","ai","image-generation","video-generation","music-generation","text-to-speech","nano-banana","gpt-image-2","veo3","suno","elevenlabs","bytedance","qwen","runway","midjourney","wan","hailuo","openai-4o-image","flux-kontext","flux-2","recraft","ideogram","topaz","model-context-protocol","ai-api","unified-ai"],"repository":{"type":"git","url":"git+https://github.com/contextarchitect/kie-ai-mcp-server.git"},"author":{"name":"contextarchitect","url":"fork of work by felores"},"bugs":{"url":"https://github.com/contextarchitect/kie-ai-mcp-server/issues"},"license":"MIT","readme":"# Kie.ai MCP Server (contextarchitect fork)\n\n**Access State of the Art AI models at half the price using one MCP Server.** Generate videos, images, music, and audio with the latest generative AI technologies using a developer-friendly API.\n\nKie.ai offers **30-50% lower cost** than competitors with 99.9% uptime and 24/7 human support.\n\n> **Fork notice**: This package (`@contextarchitect/kie-ai-mcp-server`) is a fork of\n> [`@felores/kie-ai-mcp-server`](https://github.com/felores/kie-ai-mcp-server), adding a\n> `gpt_image_2_image` tool for OpenAI's GPT Image 2 model. See `CHANGELOG.md` and `NOTICE`.\n\n## 📚 Documentation\n\n- **[Complete Tool Reference](docs/TOOLS.md)** - Detailed documentation for all AI tools (25 total, including GPT Image 2)\n- **[Database & Task Management](docs/DATABASE.md)** - SQLite database and task lifecycle  \n- **[Administrator Configuration](docs/ADMIN.md)** - Deployment guides and environment setup\n- **[Intelligent Features](docs/INTELLIGENCE.md)** - Smart mode detection and cost optimization\n\n## 🚀 Quick Start - Add to Your MCP Client\n\nThe easiest way to use this server is to add it to your MCP client configuration:\n\n```json\n{\n  \"mcpServers\": {\n    \"kie-ai\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@contextarchitect/kie-ai-mcp-server\"],\n      \"env\": {\n        \"KIE_AI_API_KEY\": \"your-api-key-here\"\n      }\n    }\n  }\n}\n```\n\n**Get your free API key:** [kie.ai/api-key](https://kie.ai/api-key)\n\n**That's it!** No callback URL setup required - the server handles it automatically.\n\n<details>\n<summary><strong>🎛️ Filter Tools (optional - click to expand)</strong></summary>\n\nReduce cognitive load by enabling only the tools you need:\n\n**Whitelist (enable specific tools only):**\n```json\n{\n  \"mcpServers\": {\n    \"kie-ai\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@contextarchitect/kie-ai-mcp-server\"],\n      \"env\": {\n        \"KIE_AI_API_KEY\": \"your-api-key-here\",\n        \"KIE_AI_ENABLED_TOOLS\": \"nano_banana_image,veo3_generate_video,suno_generate_music\"\n      }\n    }\n  }\n}\n```\n\n**Category filter (enable by type):**\n```json\n{\n  \"mcpServers\": {\n    \"kie-ai\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@contextarchitect/kie-ai-mcp-server\"],\n      \"env\": {\n        \"KIE_AI_API_KEY\": \"your-api-key-here\",\n        \"KIE_AI_TOOL_CATEGORIES\": \"image,video\"\n      }\n    }\n  }\n}\n```\n\n**Blacklist (disable specific tools):**\n```json\n{\n  \"mcpServers\": {\n    \"kie-ai\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@contextarchitect/kie-ai-mcp-server\"],\n      \"env\": {\n        \"KIE_AI_API_KEY\": \"your-api-key-here\",\n        \"KIE_AI_DISABLED_TOOLS\": \"midjourney_generate,runway_aleph_video\"\n      }\n    }\n  }\n}\n```\n\n**Categories:** `image` | `video` | `audio` — Utility tools (list_tasks, get_task_status) are always enabled.\n\n</details>\n\n**For Claude Desktop:** Add this to `~/Library/Application Support/Claude/claude_desktop_config.json` (Mac) or `%APPDATA%\\Claude\\claude_desktop_config.json` (Windows)\n\n**Works with Cursor, Windsurf, VS Code, Claude Code, OpenCode, Droid, etc.**\n\n## Why Choose Kie.ai MCP Server?\n\n| Feature | Kie.ai | Fal.ai | Replicate.com |\n|---------|--------|--------|---------------|\n| **Pricing** | 30-50% Lower | Higher | Higher |\n| **Uptime** | 99.9% | Not disclosed | Not disclosed |\n| **Support** | 24/7 Human | Email + Discord | 24/7 AI |\n| **Free Trial** | Yes | Limited | Limited |\n\n### 🚀 All AI Models in One API\n\n- **Google Veo 3**: Cinematic video generation with synchronized audio and 1080p output\n- **OpenAI Sora 2**: Advanced video generation with text/image/storyboard modes (unified)\n- **Runway Aleph**: Advanced video editing with object removal and style transfer\n- **Suno V5**: Professional music generation with realistic vocals up to 8 minutes\n- **Nano Banana 2**: Lightning-fast image generation and editing with Google Search grounding (unified tool)\n- **ElevenLabs**: Studio-quality text-to-speech and sound effects\n- **ByteDance Seedance**: High-quality video with text-to-video and image-to-video (unified)\n- **ByteDance Seedream V5 Lite**: Advanced image generation and editing with unified interface\n- **Qwen**: Powerful image generation and editing with acceleration options (unified)\n- **OpenAI 4o Image**: Advanced image generation, editing, and variant creation with GPT-4o (unified)\n- **Flux Kontext**: Professional image generation and editing with advanced features (unified)\n- **Alibaba Wan 2.5**: High-quality video generation with text-to-video and image-to-video (unified)\n- **Hailuo 02**: Professional video generation with text-to-video and image-to-video modes (unified, standard/pro quality)\n- **Kling 3.0**: Advanced video generation with 3-15s duration, multi-shot storytelling, and native audio\n- **Midjourney AI**: Industry-leading image and video generation with multiple modes (unified)\n- **Recraft Remove Background**: Professional AI-powered background removal\n- **Ideogram V3 Reframe**: Intelligent image reframing and aspect ratio conversion\n\n## What You Can Build\n\n| Category | Use Cases |\n|----------|-----------|\n| **🎬 Video Generation** | Social media content, marketing materials, product demonstrations, creative projects |\n| **🎨 Image Generation** | Content creation, product photography, artistic projects, design mockups |\n| **🎵 Music Generation** | Background music for videos, podcast intros/outros, game soundtracks, commercial projects |\n| **🎤 Audio Generation** | Narration and voiceovers, podcast production, game audio, accessibility features |\n\n## MCP Features\n\n### 🎨 Agent Prompts (Slash Commands)\n\nTrigger specialized AI agents with simple commands in your MCP client:\n\n- **`/artist`** - Image generation and editing agent  \n  Just describe what you want: _\"/artist create a logo for a coffee shop\"_\n\n- **`/filmographer`** - Video generation agent  \n  Just describe what you want: _\"/filmographer create a 10-second sunset video\"_\n\n### 📚 Knowledge Resources\n\nYour AI assistant can research and learn about available models before using them:\n\n**Agent Instructions:**\n- `kie://agents/artist` - Complete image generation workflow\n- `kie://agents/filmographer` - Complete video generation workflow\n\n**Model Documentation (33+ models):**\n- `kie://models/bytedance-seedream` - 4K image generation\n- `kie://models/veo3` - Premium cinematic video\n- `kie://models/qwen-image` - Fast image processing\n- `kie://models/flux-kontext` - Professional image generation\n- ...and 29 more models\n\n**Comparison Guides:**\n- `kie://guides/image-models-comparison` - Feature matrix for all image models\n- `kie://guides/video-models-comparison` - Feature matrix for all video models\n- `kie://guides/quality-optimization` - Cost/quality strategies\n\n### 🛠️ 21 Unified AI Tools\n\nAll tools feature **smart mode detection** - one tool does multiple things:\n\n| Category | Tools |\n|----------|-------|\n| **Image (7)** | `bytedance_seedream_image`, `qwen_image`, `nano_banana_image`, `flux_kontext_image`, `openai_4o_image`, `recraft_remove_background`, `ideogram_reframe` |\n| **Video (8)** | `veo3_generate_video`, `sora_video`, `bytedance_seedance_video`, `wan_video`, `hailuo_video`, `kling_video`, `runway_aleph_video`, `midjourney_generate` |\n| **Audio (3)** | `suno_generate_music`, `elevenlabs_tts`, `elevenlabs_ttsfx` |\n| **Utility (3)** | `list_tasks`, `get_task_status`, `veo3_get_1080p_video` |\n\n**→ [See complete tool documentation](docs/TOOLS.md)**\n\n## Key Features\n\n- **🎯 One API Key**: Access all models with one credential\n- **🤖 AI Agent Prompts**: Slash commands trigger specialized workflows\n- **📖 Knowledge Base**: 33+ resources for model research and comparison\n- **🔄 Task Management**: Built-in SQLite database for tracking generations\n- **📱 Smart Routing**: Automatic endpoint detection and status monitoring\n- **🛡️ Error Handling**: Validation and error recovery\n- **⚙️ Flexible Parameters**: Control outputs with parameters\n- **📊 Persistent Storage**: Tasks survive server restarts\n- **🎛️ Quality Control**: Choose between speed (lite) and quality (pro) modes\n- **🌐 Multilingual Support**: Text-to-speech in multiple languages\n\n## 🧠 Intelligent Intention Detection System\n\nThe MCP server features advanced **intention detection algorithms** that automatically understand user requirements and optimize both cost and quality without manual configuration.\n\n### Quick Summary\n\n- **Automatic Quality Detection**: Analyzes user language (\"high quality\" → pro models, \"quick\" → lite models)\n- **Smart Mode Detection**: Single tools auto-detect operation mode (generate/edit/upscale) based on parameters\n- **Database-Driven Intelligence**: Local SQLite cache reduces API calls and provides smart routing\n- **Cost Control by Design**: Defaults to cheapest options (720p, lite quality) unless explicitly requested\n\n**Example**: User says _\"Make a quick social media video\"_ → System automatically chooses: lite quality + 720p + 5 second duration = lowest cost tier (1x baseline)\n\n**Example**: User says _\"I need a high quality video for a client presentation\"_ → System automatically chooses: pro quality + 1080p = highest cost tier (4-6x baseline)\n\n**→ [See complete intelligence documentation](docs/INTELLIGENCE.md)** with real-world examples and verifiable code references\n\n## Installation & Configuration\n\n<details>\n<summary><strong>📦 Installation Options (click to expand)</strong></summary>\n\n### Option 1: Install from NPM (Recommended)\n```bash\nnpm install -g @contextarchitect/kie-ai-mcp-server\n```\n\n### Option 2: Install from Source\n```bash\n# Clone the repository\ngit clone https://github.com/felores/kie-ai-mcp-server.git\ncd kie-ai-mcp-server\n\n# Install dependencies\nnpm install\n\n# Build the project\nnpm run build\n```\n</details>\n\n<details>\n<summary><strong>⚙️ Environment Variables (click to expand)</strong></summary>\n\n### Required\n```bash\nexport KIE_AI_API_KEY=\"your-api-key-here\"  # Get from https://kie.ai/api-key\n```\n\n### Optional\n```bash\nexport KIE_AI_BASE_URL=\"https://api.kie.ai/api/v1\"  # Default API base URL\nexport KIE_AI_TIMEOUT=\"60000\"                       # Request timeout (ms)\nexport KIE_AI_DB_PATH=\"./tasks.db\"                  # Database file location\nexport KIE_AI_CALLBACK_URL=\"https://your-domain.com/webhook\"  # Custom callback\nexport KIE_AI_CALLBACK_URL_FALLBACK=\"https://your-proxy.com/callback\"  # Admin fallback\n```\n\n### Callback URL Priority\n\n| Priority | Source | Variable | Use Case |\n|----------|--------|----------|----------|\n| 1 | User Parameter | `callBackUrl` | Per-request override |\n| 2 | Environment | `KIE_AI_CALLBACK_URL` | User's custom callback |\n| 3 | Admin Fallback | `KIE_AI_CALLBACK_URL_FALLBACK` | ⭐ Deployment-wide default |\n| 4 | Hardcoded | - | `https://proxy.kie.ai/mcp-callback` |\n\n**→ [See administrator configuration guide](docs/ADMIN.md)** for Docker, Kubernetes, Systemd examples\n</details>\n\n### Tool Filtering (v2.0.2+)\n\n**Filter which AI tools are available** to reduce cognitive load and focus your workflow:\n\n```bash\n# Whitelist: Enable only specific tools (highest priority)\n# Note: Utility tools (list_tasks, get_task_status) are always included automatically\nexport KIE_AI_ENABLED_TOOLS=\"nano_banana_image,veo3_generate_video,suno_generate_music\"\n\n# Category filter: Enable by category (medium priority)\nexport KIE_AI_TOOL_CATEGORIES=\"image,video\"  # Categories: image, video, audio\n\n# Blacklist: Disable specific tools (lowest priority)\n# Note: Utility tools cannot be disabled\nexport KIE_AI_DISABLED_TOOLS=\"midjourney_generate,runway_aleph_video\"\n```\n\n**Priority Logic**: `ENABLED_TOOLS` > `TOOL_CATEGORIES` > `DISABLED_TOOLS` > All tools (default)\n\n**Tool Categories**:\n- **image** (8): nano_banana, seedream, qwen, openai_4o, flux, recraft, ideogram, midjourney*\n- **video** (9): veo3, veo3_1080p, sora, seedance, wan, hailuo, kling, runway, midjourney*\n- **audio** (3): suno, elevenlabs_tts, elevenlabs_ttsfx\n- **utility** (2): list_tasks, get_task_status ⭐ **Always enabled**\n\n_* midjourney appears in both image and video categories (supports both)_\n- ⭐ **Utility tools are always enabled** for server monitoring and task management\n- When using whitelist mode, utility tools are automatically added to your selection\n- When using blacklist mode, utility tools cannot be disabled (warning shown if attempted)\n\n<details>\n<summary><strong>🔧 MCP Client Configuration (click to expand)</strong></summary>\n\n### Claude Desktop, Cursor, Windsurf, VS Code, etc.\n\nAdd to your MCP client configuration file:\n\n```json\n{\n  \"kie-ai-mcp-server\": {\n    \"command\": \"node\",\n    \"args\": [\"/path/to/kie-ai-mcp-server/dist/index.js\"],\n    \"env\": {\n      \"KIE_AI_API_KEY\": \"your-api-key-here\"\n    }\n  }\n}\n```\n\nOr if installed globally with npx:\n\n```json\n{\n  \"kie-ai-mcp-server\": {\n    \"command\": \"npx\",\n    \"args\": [\"-y\", \"@contextarchitect/kie-ai-mcp-server\"],\n    \"env\": {\n      \"KIE_AI_API_KEY\": \"your-api-key-here\"\n    }\n  }\n}\n```\n</details>\n\n## Quick Examples\n\n### Generate Image\n```json\n{\n  \"tool\": \"nano_banana_image\",\n  \"arguments\": {\n    \"prompt\": \"A futuristic city at sunset, cyberpunk style\",\n    \"image_size\": \"16:9\",\n    \"output_format\": \"png\"\n  }\n}\n```\n\n### Generate Video\n```json\n{\n  \"tool\": \"sora_video\",\n  \"arguments\": {\n    \"prompt\": \"A peaceful garden with blooming flowers and butterflies\",\n    \"model\": \"sora-2\",\n    \"resolution\": \"1080p\",\n    \"duration\": \"10\"\n  }\n}\n```\n\n### Generate Music\n```json\n{\n  \"tool\": \"suno_generate_music\",\n  \"arguments\": {\n    \"prompt\": \"Upbeat electronic music with energetic beats\",\n    \"customMode\": true,\n    \"instrumental\": false,\n    \"model\": \"V5\",\n    \"style\": \"Electronic\",\n    \"title\": \"Energy Boost\"\n  }\n}\n```\n\n### Text-to-Speech\n```json\n{\n  \"tool\": \"elevenlabs_tts\",\n  \"arguments\": {\n    \"text\": \"Welcome to the future of AI-powered content creation!\",\n    \"voice\": \"Rachel\",\n    \"model\": \"turbo\"\n  }\n}\n```\n\n**→ [See 100+ more examples in tool documentation](docs/TOOLS.md)**\n\n## Database & Task Management\n\nThe server includes a built-in SQLite database for persistent task tracking:\n\n- **🔄 Persistent Storage**: Tasks survive server restarts\n- **📊 Complete History**: Track all generation tasks and their results\n- **⚡ Smart Caching**: Local database reduces API calls\n- **🔍 Full Audit Trail**: Complete lifecycle tracking\n- **🎯 Intelligent Routing**: Database provides api_type for correct endpoint selection\n\n### Quick Examples\n\n**List recent tasks:**\n```json\n{\n  \"tool\": \"list_tasks\",\n  \"arguments\": {\n    \"limit\": 20,\n    \"status\": \"completed\"\n  }\n}\n```\n\n**Check task status:**\n```json\n{\n  \"tool\": \"get_task_status\",\n  \"arguments\": {\n    \"task_id\": \"281e5b0*********************f39b9\"\n  }\n}\n```\n\n**→ [See complete database documentation](docs/DATABASE.md)** including schema, lifecycle, and best practices\n\n## Real-World Use Cases\n\n<details>\n<summary><strong>🎬 Content Creation Agencies (click to expand)</strong></summary>\n\n```bash\n# Generate social media video content\nsora_video: \"A trendy coffee shop with latte art, cinematic lighting\"\n\n# Create product photography\nnano_banana_image: \"Luxury watch on marble surface, professional product shot\"\n\n# Add background music\nsuno_generate_music: \"Upbeat corporate background music, 2 minutes\"\n```\n</details>\n\n<details>\n<summary><strong>🎮 Game Development Studios (click to expand)</strong></summary>\n\n```bash\n# Generate game assets\nbytedance_seedream_image: \"Fantasy sword with glowing runes, game asset style\"\n\n# Create character voiceovers\nelevenlabs_tts: \"Welcome, brave adventurer! Your quest begins now.\"\n\n# Design sound effects\nelevenlabs_ttsfx: \"Magical spell casting with sparkles and energy\"\n```\n</details>\n\n<details>\n<summary><strong>📱 Mobile App Developers (click to expand)</strong></summary>\n\n```bash\n# Generate app icons and illustrations\nflux_kontext_image: \"Modern minimalist app icon for fitness tracker\"\n\n# Create tutorial videos\nbytedance_seedance_video: \"Screen recording showing app features, clean interface\"\n\n# Add narration\nelevenlabs_tts: \"Tap here to get started with your new profile\"\n```\n</details>\n\n<details>\n<summary><strong>🏢 Enterprise Applications (click to expand)</strong></summary>\n\n```bash\n# Generate training materials\nveo3_generate_video: \"Professional office environment, employee training scenario\"\n\n# Create corporate presentations\nopenai_4o_image: \"Add company logo to presentation slide, maintain professional style\"\n\n# Produce marketing content\nsuno_generate_music: \"Corporate background music for promotional video\"\n```\n</details>\n\n## Error Handling\n\nThe server handles these HTTP error codes from Kie.ai:\n\n| Code | Meaning |\n|------|---------|\n| **200** | Success |\n| **400** | Content policy violation / English prompts only |\n| **401** | Unauthorized (invalid API key) |\n| **402** | Insufficient credits |\n| **404** | Resource not found |\n| **422** | Validation error / record is null |\n| **429** | Rate limited |\n| **451** | Image access limits |\n| **455** | Service maintenance |\n| **500** | Server error / timeout |\n| **501** | Generation failed |\n\n## Development\n\n```bash\n# Run tests\nnpm test\n\n# Development mode with auto-reload\nnpm run dev\n\n# Type checking\nnpx tsc --noEmit\n\n# Build for production\nnpm run build\n```\n\n## Pricing\n\nBased on Kie.ai documentation:\n- **Nano Banana**: $0.020 per image (4 credits)\n- **Veo3 Quality**: Higher cost tier\n- **Veo3 Fast**: ~20% of Quality model pricing\n\nSee https://kie.ai/billing for detailed pricing.\n\n## Production Tips\n\n1. **Database Location**: Set `KIE_AI_DB_PATH` to a persistent location\n2. **API Key Security**: Never commit API keys to version control\n3. **Rate Limiting**: Implement client-side rate limiting for high-volume usage\n4. **Monitoring**: Monitor task status and handle failed generations appropriately\n5. **Storage**: Consider automatic cleanup of old task records\n\n**→ [See complete administrator guide](docs/ADMIN.md)** for deployment best practices\n\n## Troubleshooting\n\n### Common Issues\n\n**\"Unauthorized\" errors**\n- Verify `KIE_AI_API_KEY` is set correctly\n- Check API key is valid at https://kie.ai/api-key\n\n**\"Task not found\" errors**\n- Tasks may expire after 14 days\n- Check task ID format matches expected pattern\n\n**Generation failures**\n- Check content policy compliance\n- Verify prompt is in English\n- Ensure sufficient API credits\n\n## Support\n\nFor issues related to:\n- **MCP Server**: Open an issue at https://github.com/felores/kie-ai-mcp-server/issues\n- **Kie.ai API**: Contact support@kie.ai or check https://docs.kie.ai/\n- **API Keys**: Visit https://kie.ai/api-key\n\n## 🚀 Start Building with Kie.ai\n\n### 🎯 Get Started\n1. **Get your free API key** at [kie.ai/api-key](https://kie.ai/api-key)\n2. **Install the MCP server**: `npm install @contextarchitect/kie-ai-mcp-server`\n3. **Generate your first AI content** in minutes\n\n### 💡 Benefits\n- ✅ **Free trial** - Test models before paying\n- ✅ **30-50% lower pricing** than competitors\n- ✅ **99.9% uptime** guarantee\n- ✅ **24/7 human support**\n- ✅ **Simple integration**\n\n### 🌟 AI Content Generation\nKie.ai provides access to advanced AI models at competitive pricing.\n\n**Start your project today.** 🚀\n\n---\n\n## License\n\nMIT License - see LICENSE file for details.\n\n## Contributing\n\n1. Fork the repository\n2. Create a feature branch\n3. Make your changes\n4. Add tests if applicable\n5. Submit a pull request\n\n## Changelog\n\nSee [CHANGELOG.md](CHANGELOG.md) for detailed version history and release notes.\n","readmeFilename":"README.md","_rev":"1-c909471cbf09a9e2e63e79e03ee6e683"}