KieAI MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| PORT | No | Port for HTTP/SSE mode | 3001 |
| KIE_API_KEY | Yes | API key from kie.ai (required) | |
| MCP_TRANSPORT | No | Transport mode: 'stdio' or 'http' | stdio |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| kie_generate_imageB | Generate an image using AI models via kie.ai |
| kie_generate_videoC | Generate a video using AI models via kie.ai |
| kie_generate_audioA | Generate audio (text-to-speech) using AI models via kie.ai |
| kie_upscaleB | Upscale an image using AI models via kie.ai |
| kie_check_taskB | Check the status of a kie.ai generation task |
| kie_list_modelsA | List available AI models from the kie.ai registry |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 6 tools
Each tool targets a distinct operation: generating images, videos, audio, checking task status, upscaling images, and listing models. There is no overlap or ambiguity between these actions.
All tool names follow a consistent kie_ prefix with snake_case verbs, but 'kie_upscale' is a verb alone while others use verb_noun (e.g., kie_generate_image). This minor deviation keeps it mostly predictable.
Six tools are well-scoped for an AI media generation server, covering generation across modalities plus task management and model discovery without unnecessary bloat.
The surface covers generation, upscaling, task checking, and model listing, which forms a cohesive workflow. Minor gaps like explicit cancelation or result retrieval are indirectly handled via check_task, so the core scope is complete.