kokon
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| QLS_MCP_ALLOWED_DIRS | No | Folders the MCP server is allowed to read from and write to, in addition to the project folder and Kokon's output folder. Multiple folders are separated by semicolons. |
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 |
|---|---|
| generate_imageA | Generate an image from a text prompt with the local Qwen-Image 2.1 model, optionally using reference images (a person, a product, a style) in a new scene. Write the full, detailed prompt yourself (see the prompt field): it is used as is. Dimensions are chosen automatically (aspect ratio can be inferred from the prompt, e.g. '16:9 hero image') unless width/height or aspect_ratio are given. Returns the absolute path of the saved PNG. Runs fully offline; typically 45 s–4 min. |
| edit_imageA | Edit an existing image with a natural-language instruction (e.g. 'make the background nighttime, keep the foreground unchanged'). Preserves the input's width, height and aspect ratio by default. The input file is never overwritten. |
| edit_imagesA | Edit image 1 with the help of other images (up to 10 in all) and an instruction, e.g. put the product from image 2 in the hand of the person in image 1. Order matters: refer to them as image 1, image 2, ... in the prompt. The output follows image 1's aspect ratio unless width/height or aspect_ratio are given. For a new scene that only uses the images as references, use generate_image with reference_images. |
| outpaint_imageA | Extend an image beyond its borders. Give the expansion per side in pixels or percent; the canvas, mask and model-aligned sizes are built automatically and the original pixels are preserved exactly. |
| inpaint_imageA | Change only the masked area of an image (mask: white = change, black = keep): edit it, remove what is there, replace it, or add something inside it. The model is shown the area marked on a copy of the image; pixels outside the mask are preserved exactly. |
| upscale_imageA | Enlarge an image (default 2x, or to a width/height up to 8192 px). Runs the upscaler model chosen in Kokon (Superscale-SP by default; seconds on a PC GPU, longer on a Mac; adds no new content). Saves a new file; the input is never overwritten. |
| get_image_infoC | Width, height, aspect ratio, file type and, for images made by Kokon, the generation metadata (prompt, seed, model, steps...). |
| get_generator_statusB | Whether the local inference server is running and the model loaded, GPU and VRAM, the active job and the queue. |
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 8 tools
Each tool maps to a distinct operation (generate, single-image edit, multi-image edit, outpaint, upscale, mask inpaint, info, status), and the descriptions explicitly disambiguate the trickiest pair by telling the agent when to prefer generate_image with reference_images over edit_images. The only residual overlap is edit_image vs inpaint_image vs edit_images, which share the 'modify an image' space but differ in inputs.
All names are snake_case verb_noun following a predictable pattern (generate_image, edit_image, edit_images, outpaint_image, upscale_image, inpaint_image, get_image_info). The sole deviation is get_generator_status, whose noun ('generator') breaks the image-centric pattern, and edit_image/edit_images differ only by plural.
Eight tools sit squarely in the well-scoped range, and each covers a genuinely different image workflow (generate, edit, multi-reference edit, outpaint, upscale, inpaint, inspect, status). No redundant or filler tools are present.
The lifecycle is well covered — generation, several editing modalities, upscaling, and metadata inspection. Minor gaps: status reports an active job and queue but there is no cancel/queue-listing tool, and there is no way to list or manage previously generated files; agents can work around both via absolute paths and get_image_info.