Connect AI agents to the Recraft AI MCP server

1. Install the SDK

   Node.js

   npm install @scalekit-sdk/node dotenv

   Python

   pip install scalekit-sdk-python python-dotenv

   Full SDK reference: Node.js | Python

2. Set your credentials

   Add your Scalekit credentials to your .env file. Find values in app.scalekit.com > Developers > API Credentials.

   .env

   SCALEKIT_ENVIRONMENT_URL=<your-environment-url>
   SCALEKIT_CLIENT_ID=<your-client-id>
   SCALEKIT_CLIENT_SECRET=<your-client-secret>

3. Authorize and make your first call

   Node.js

   quickstart.mts

   import { ScalekitClient } from '@scalekit-sdk/node'
   import 'dotenv/config'
   import { createInterface } from 'node:readline/promises'
   
   const scalekit = new ScalekitClient(
     process.env.SCALEKIT_ENVIRONMENT_URL,
     process.env.SCALEKIT_CLIENT_ID,
     process.env.SCALEKIT_CLIENT_SECRET,
   )
   const actions = scalekit.actions
   
   const connector = 'recraftmcp'
   const identifier = 'user_123'
   
   // Generate an authorization link for the user
   const { link } = await actions.getAuthorizationLink({ connectionName: connector, identifier })
   console.log('Authorize Recraft AI MCP:', link)
   const rl = createInterface({ input: process.stdin, output: process.stdout })
   await rl.question('Press Enter after authorizing...')
   rl.close()
   
   // Make your first call
   const result = await actions.executeTool({
     connector,
     identifier,
     toolName: 'recraftmcp_get_user',
     toolInput: {},
   })
   console.log(result)

   npx tsx quickstart.mts

   Python

   quickstart.py

   import os
   from scalekit import ScalekitClient
   from dotenv import load_dotenv
   load_dotenv()
   
   scalekit_client = ScalekitClient(
       env_url=os.getenv("SCALEKIT_ENVIRONMENT_URL"),
       client_id=os.getenv("SCALEKIT_CLIENT_ID"),
       client_secret=os.getenv("SCALEKIT_CLIENT_SECRET"),
   )
   actions = scalekit_client.actions
   
   connection_name = "recraftmcp"
   identifier = "user_123"
   
   # Generate an authorization link for the user
   link_response = actions.get_authorization_link(
       connection_name=connection_name,
       identifier=identifier,
   )
   print("Authorize Recraft AI MCP:", link_response.link)
   input("Press Enter after authorizing...")
   
   # Make your first call
   result = actions.execute_tool(
       tool_input={},
       tool_name="recraftmcp_get_user",
       connection_name=connection_name,
       identifier=identifier,
   )
   print(result)

   python quickstart.py

What you can do

Connect this agent connector to let your agent:

• Edit image — Edit one or more input images according to a text prompt, producing new image(s)
• Agent call — Talk to Recraft’s Design Agent in a multi-turn chat to turn a brief for a digital product into a coherent set of design assets (logo, colour palette, typography, app icon, social assets) and optionally save them as a reusable Design Kit
• Image vectorize, variate, inpaint — Convert a raster image to a vector format
• Model suggest — Suggest the best Recraft image generation model for a given user request
• Plans subscription — List available Recraft subscription plans with their credits, refill periods, and pricing
• Url request upload — Issue an upload URL for a direct image upload

Tool list

Use the exact tool names from the Tool list below when you call execute_tool. If you’re not sure which name to use, list the tools available for the current user first.

recraftmcp_call_agent#Talk to Recraft's Design Agent in a multi-turn chat to turn a brief for a digital product into a coherent set of design assets (logo, colour palette, typography, app icon, social assets) and optionally save them as a reusable Design Kit. The agent may reply with a clarifying question instead of a final result — relay its reply to the end user verbatim and send their answer back in a follow-up call rather than answering on its behalf.3 params▾

Talk to Recraft's Design Agent in a multi-turn chat to turn a brief for a digital product into a coherent set of design assets (logo, colour palette, typography, app icon, social assets) and optionally save them as a reusable Design Kit. The agent may reply with a clarifying question instead of a final result — relay its reply to the end user verbatim and send their answer back in a follow-up call rather than answering on its behalf.

NameTypeRequiredDescription

messagestringrequiredText message to send to the Recraft agent.

chat_idstringoptionalUUID of an existing chat to continue. Omit to start a new chat.

input_image_urlsarrayoptionalPublicly accessible HTTPS URLs of input images to send to the agent (1–10).

recraftmcp_create_style#Create a custom style from one or more reference images. Provide images via URLs or base64-encoded data. The style parameter defines the base style type.6 params▾

Create a custom style from one or more reference images. Provide images via URLs or base64-encoded data. The style parameter defines the base style type.

NameTypeRequiredDescription

input_image_typestringrequiredVisual image type for the style

input_style_images_b64arrayoptionalBase64-encoded reference images (1–5). Mutually exclusive with input_style_images_urls.

input_style_images_urlsarrayoptionalPublicly accessible HTTPS URLs of reference images (1–5). Mutually exclusive with input_style_images_b64.

modelstringoptionalModel version. Default: recraftv4_1.

privatebooleanoptionalWhether the style is private.

style_promptstringoptionalOptional text prompt to refine style (1–1000 chars).

recraftmcp_creative_upscale#Upscale an image using creative AI enhancement. Returns the upscaled image as a URL and a WEBP preview.1 param▾

Upscale an image using creative AI enhancement. Returns the upscaled image as a URL and a WEBP preview.

NameTypeRequiredDescription

input_image_urlstringrequiredPublicly accessible HTTPS URL of the input image (PNG/JPEG/WEBP). Use request_upload_url first for local files.

recraftmcp_crisp_upscale#Upscale an image with sharp, crisp quality enhancement. Returns the upscaled image as a URL and a WEBP preview.1 param▾

Upscale an image with sharp, crisp quality enhancement. Returns the upscaled image as a URL and a WEBP preview.

NameTypeRequiredDescription

input_image_urlstringrequiredPublicly accessible HTTPS URL of the input image (PNG/JPEG/WEBP). Use request_upload_url first for local files.

recraftmcp_delete_style#Delete a custom style by its ID.1 param▾

Delete a custom style by its ID.

NameTypeRequiredDescription

input_style_idstringrequiredID of the style to delete.

recraftmcp_erase_region#Erase a masked region from an image, filling it with content-aware background. Returns the processed image as a URL and a WEBP preview.2 params▾

Erase a masked region from an image, filling it with content-aware background. Returns the processed image as a URL and a WEBP preview.

NameTypeRequiredDescription

input_image_urlstringrequiredPublicly accessible HTTPS URL of the source image (PNG/JPEG/WEBP).

mask_image_urlstringrequiredPublicly accessible HTTPS URL of the mask image. White areas will be erased.

recraftmcp_generate_background#Generate a background for a masked region of an image based on a text prompt. Returns the processed image as a URL and a WEBP preview.7 params▾

Generate a background for a masked region of an image based on a text prompt. Returns the processed image as a URL and a WEBP preview.

NameTypeRequiredDescription

input_image_urlstringrequiredPublicly accessible HTTPS URL of the source image (PNG/JPEG/WEBP).

mask_image_urlstringrequiredPublicly accessible HTTPS URL of the mask image. White areas define the background region.

promptstringrequiredText prompt for the background (1–10000 chars).

input_stylestringoptionalVisual style. Mutually exclusive with input_style_id.

input_style_idstringoptionalStyle ID to apply. Mutually exclusive with input_style.

modelstringoptionalModel version.

nintegeroptionalNumber of images to generate (1–6). Default: 1.

recraftmcp_generate_image#Generate an image from a text prompt. Returns image URLs and WEBP previews.6 params▾

Generate an image from a text prompt. Returns image URLs and WEBP previews.

NameTypeRequiredDescription

promptstringrequiredText prompt (1–10000 chars).

image_sizestringoptionalImage dimensions or aspect ratio. Default: 1:1.

input_stylestringoptionalVisual style. Mutually exclusive with input_style_id.

input_style_idstringoptionalStyle ID. Mutually exclusive with input_style.

modelstringoptionalModel version. Default: recraftv4_1.

nintegeroptionalNumber of images (1–6). Default: 1.

recraftmcp_get_style#Get details of a specific style by its ID.1 param▾

Get details of a specific style by its ID.

NameTypeRequiredDescription

style_idstringrequiredUUID of the style (e.g. '550e8400-e29b-41d4-a716-446655440000').

recraftmcp_get_user#Get information about the current user including ID, email, name, and credit balance.0 params▾

Get information about the current user including ID, email, name, and credit balance.

recraftmcp_image_edit#Edit one or more input images according to a text prompt, producing new image(s). Use for instruction-driven editing (e.g. "add a hat", "combine these images") where the model follows explicit editing instructions or uses an external model; use image_to_image instead for a strength-controlled stylistic transformation of a single image.6 params▾

Edit one or more input images according to a text prompt, producing new image(s). Use for instruction-driven editing (e.g. "add a hat", "combine these images") where the model follows explicit editing instructions or uses an external model; use image_to_image instead for a strength-controlled stylistic transformation of a single image.

NameTypeRequiredDescription

input_image_urlsarrayrequiredPublicly accessible HTTPS URLs of the source images to edit (1–10).

promptstringrequiredText prompt describing the edit to apply (1–10000 chars).

image_sizestringoptionalImage dimensions or aspect ratio. Defaults to the first input image's size if omitted.

input_style_idstringoptionalID of the style to apply. Mutually exclusive with style.

modelstringoptionalModel version to use. Default: recraftv4_1.

nintegeroptionalNumber of images to generate (1–6). Default: 1.

recraftmcp_image_to_image#Transform an existing image based on a text prompt. The strength parameter controls how much the output differs from the input.8 params▾

Transform an existing image based on a text prompt. The strength parameter controls how much the output differs from the input.

NameTypeRequiredDescription

input_image_urlstringrequiredPublicly accessible HTTPS URL of the input image (PNG/JPEG/WEBP).

promptstringrequiredText prompt for the transformation (1–10000 chars).

image_sizestringoptionalOutput image dimensions.

input_stylestringoptionalVisual style. Mutually exclusive with input_style_id.

input_style_idstringoptionalStyle ID. Mutually exclusive with input_style.

modelstringoptionalModel version.

nintegeroptionalNumber of images (1–6).

strengthnumberoptionalTransformation strength (0=identical, 1=fully transformed). Default: 0.6.

recraftmcp_inpaint_image#Fill in a masked region of an image based on a text prompt. Returns the processed image as a URL and a WEBP preview.7 params▾

Fill in a masked region of an image based on a text prompt. Returns the processed image as a URL and a WEBP preview.

NameTypeRequiredDescription

input_image_urlstringrequiredPublicly accessible HTTPS URL of the source image (PNG/JPEG/WEBP).

mask_image_urlstringrequiredPublicly accessible HTTPS URL of the mask image. White areas define the inpainting region.

promptstringrequiredText prompt describing what to paint in the masked region (1–10000 chars).

input_stylestringoptionalVisual style. Mutually exclusive with input_style_id.

input_style_idstringoptionalStyle ID. Mutually exclusive with input_style.

modelstringoptionalModel version to use for generation.

nintegeroptionalNumber of images to generate (1–6).

recraftmcp_list_styles#List all custom styles created by the current user.0 params▾

List all custom styles created by the current user.

recraftmcp_remove_background#Remove the background from an image. Returns the result as a URL and a WEBP preview.1 param▾

Remove the background from an image. Returns the result as a URL and a WEBP preview.

NameTypeRequiredDescription

input_image_urlstringrequiredPublicly accessible HTTPS URL of the image (PNG/JPEG/WEBP).

recraftmcp_replace_background#Replace the background of an image based on a text prompt. Returns the processed image as a URL and a WEBP preview.6 params▾

Replace the background of an image based on a text prompt. Returns the processed image as a URL and a WEBP preview.

NameTypeRequiredDescription

input_image_urlstringrequiredPublicly accessible HTTPS URL of the source image (PNG/JPEG/WEBP).

promptstringrequiredText prompt describing the new background (1–10000 chars).

input_stylestringoptionalVisual style. Mutually exclusive with input_style_id.

input_style_idstringoptionalStyle ID. Mutually exclusive with input_style.

modelstringoptionalModel version to use for generation.

nintegeroptionalNumber of images to generate (1–6).

recraftmcp_request_upload_url#Issue an upload URL for a direct image upload. Use this when you have a local image file and need a publicly accessible URL. PUT the image bytes to the returned upload URL, then use the resulting image_url in other tools.0 params▾

Issue an upload URL for a direct image upload. Use this when you have a local image file and need a publicly accessible URL. PUT the image bytes to the returned upload URL, then use the resulting image_url in other tools.

recraftmcp_subscription_plans#List available Recraft subscription plans with their credits, refill periods, and pricing.0 params▾

List available Recraft subscription plans with their credits, refill periods, and pricing.

recraftmcp_suggest_model#Suggest the best Recraft image generation model for a given user request.1 param▾

Suggest the best Recraft image generation model for a given user request.

NameTypeRequiredDescription

promptstringrequiredThe user's image generation request to evaluate.

recraftmcp_variate_image#Generate variations of an existing image. Returns image URLs and WEBP previews.6 params▾

Generate variations of an existing image. Returns image URLs and WEBP previews.

NameTypeRequiredDescription

input_image_urlstringrequiredPublicly accessible HTTPS URL of the source image (PNG/JPEG/WEBP).

image_sizestringoptionalOutput image size as aspect ratio.

input_stylestringoptionalVisual style. Mutually exclusive with input_style_id.

input_style_idstringoptionalStyle ID. Mutually exclusive with input_style.

modelstringoptionalModel version to use for generation.

nintegeroptionalNumber of variations to generate (1–6).

recraftmcp_vectorize_image#Convert a raster image to a vector format. Returns the vector image as a URL.1 param▾

Convert a raster image to a vector format. Returns the vector image as a URL.

NameTypeRequiredDescription

input_image_urlstringrequiredPublicly accessible HTTPS URL of the raster image (PNG/JPEG/WEBP).