Built for AI agents
An image search MCP server for your agent.
Pix4Less is a remote Model Context Protocol server at https://pix4less.com/mcp (Streamable HTTP). Connect Claude, ChatGPT, Cursor or another MCP client, sign in, and it can search the catalog, look at previews, and download the image you choose.
Start with 100 credits. No payment is required for signup; searches and new downloads consume credits.
Setup
Connect in one step.
Add the server URL to your AI app and sign in with your Pix4Less account when it asks. There is no API key to copy. Each tool calls the same /v1 API, so prices, limits and errors match the JSON API.
Claude, ChatGPT, Cursor, VS Code, Codex, Gemini CLI and Antigravity each have their own steps, and there is a prompt that lets your agent install the server itself. Both are in the MCP docs. For a walkthrough from setup to first download, read Add image search to Claude and Cursor with MCP. Building a website instead? See Add stock images to a Next.js site.
https://pix4less.com/mcpclaude mcp add --transport http pix4less https://pix4less.com/mcp{
"mcpServers": {
"pix4less": {
"url": "https://pix4less.com/mcp"
}
}
}Scripts, CI and clients that cannot sign in can send an API key from the dashboard as an Authorization: Bearer header instead.
18 tools
What your agent can do.
Search finds existing catalog images and never generates one. Generation is a separate tool your agent calls only when nothing fits.
search_imagesSearch the catalog by description, with aspect ratio, resolution, orientation, collection, tag, size and format filters.
browse_imagesBrowse the newest images, 12 per page, without a query. Free.
random_imageDraw one random published image. Free, within a monthly allowance.
get_imageGet one image's details: title, description, tags, size, quality tier, download price, license and 4K availability. Free.
list_collectionsList published collections, to narrow a search to one collection. Free.
download_imageGet a short-lived signed URL for the full-resolution file, charged once per image per account.
generate_imageCreate a new image from a description when the catalog has nothing suitable.
get_generationCheck a generation job's status, optionally waiting for the next change. Free.
list_generationsList this account's generation jobs, newest first. Free.
list_my_imagesList the images this account owns (generated or downloaded), newest first, to find and reuse them. Free.
cancel_generationCancel a pending or generating job and refund its charge.
create_batchQueue 1-100 images at batch prices; each is delivered within 24 hours or refunded.
get_batchProgress and status of a batch. Free.
get_batch_resultsPer-item results of a batch, ready for download_image. Free.
list_presetsList the built-in and custom style presets for generate_image and create_batch. Free.
create_presetSave a custom style, such as a brand look, to reuse across generations. Free.
delete_presetDelete one of your custom presets.
get_pricingRead the live price book: search, download and generation prices. Free.
Live price book
What it costs.
There is no separate MCP fee. Tools draw on the same credits as the API, so the prices below are live from the price book.
- Quick search
- 1 credit / request
- Thorough search
- 3 credits / request
- New catalog download
- 1 credit / image (HD: 3 credits, 4K: 10 credits)
- Previously purchased image
- No duplicate download charge
- Browsing, details, collections
- Free
Searches charge only after a valid search returns at least one result. Search and download are separate charges. Image generation is priced per model and resolution.
Plans, top-ups, and generation pricingGet started.
- Add the server to your AI app with the button, command or config above, or paste the setup prompt from the MCP docs into your agent.
- Sign in with your Pix4Less account when the app asks, or create one on the way, and allow the connection.
- Ask your agent to find an image, for example a landscape photo of a coffee shop, and to show you the previews.
Images are AI-generated. Review each image and its license for your use case; the server does not guarantee a match for every query.
