Build with AI

Connect AI coding agents and assistants to ImageKit with skills, plugins, and MCP servers to manage media and build transformations using natural language.
Large language models were trained before the latest version of ImageKit, so on their own they often suggest outdated API signatures, invent transformation parameters, or pick the wrong approach. The tools on this page fix that. They give your AI assistant accurate, up-to-date knowledge of ImageKit and the ability to act on your account.
There are three ways to use AI with ImageKit:
If you are building in an IDE, the quickest way to get everything is to install the ImageKit skills, then add the MCP servers for your client.
ImageKit skills teach your AI coding agent how to use ImageKit's MCP servers correctly. Each skill is a playbook for a task the tool schema does not make obvious — searching the library, uploading, sharing, building a transformation, or querying the docs.
Every IDE installs the skills with the same command:
Then add the MCP servers shown for your tool. DAM and Admin require you to sign in to ImageKit and grant access the first time you connect. DevTools needs no authentication.
Inside Claude Code, add the marketplace and install the plugin, which bundles the skills and MCP servers:
Restart Claude Code for the MCP servers to take effect. Authenticate imagekit_dam and imagekit_admin when prompted.
Manual (Claude Desktop only)
Edit your config file:
Restart Claude Desktop for the changes to take effect. Claude.ai (web) has no local config file, so use the plugin method there.
Add the MCP servers from the skills repository's install buttons, or edit ~/.cursor/mcp.json:
Restart Cursor for the changes to take effect. Authenticate DAM and Admin when prompted.
Then edit ~/.codeium/windsurf/mcp_config.json:
Restart Windsurf for the changes to take effect. Authenticate DAM and Admin when prompted.
Or add them via the CLI:
Or edit ~/.codex/config.toml:
Restart Codex for the MCP servers to take effect. Authenticate DAM and Admin when prompted.
The Model Context Protocol (MCP) is an open standard that lets AI assistants securely connect to external tools and data. ImageKit provides hosted, remote MCP servers, so there is no local server to run or maintain. Add them to any MCP client using the URLs below.
If you previously connected https://api-mcp.imagekit.io/mcp, remove that server from your client and add DAM and Admin instead.
The DAM MCP server lets an AI assistant work in your ImageKit media library on your behalf, with the same access as your user account. Connect to it at https://imagekit.io/mcp/dam.
In supporting clients, search, upload, and file details can open an interactive UI with thumbnails and a file picker instead of plain JSON.
On first connect, you sign in to ImageKit and choose what the assistant is allowed to do. See security recommendations.
The Admin MCP server lets an AI assistant manage account-level delivery settings and webhooks, and inspect usage. Connect to it at https://imagekit.io/mcp/admin.
On first connect, you sign in to ImageKit and grant access.
The DevTools MCP server makes any AI assistant better at ImageKit, even before you connect your account. It is public and needs no authentication, so you can connect to it at https://devtools-mcp.imagekit.io/mcp. It exposes two tools:
The DAM and Admin MCP servers are hosted and maintained by ImageKit. You sign in with your ImageKit account; the assistant never sees your password or API keys.
What the assistant can do is limited by:
Grant only the access you need. View is enough for search and browse. Creating, updating, and deleting are required for uploads, edits, sharing, and destructive operations.
On first connect, ImageKit shows a consent screen so you can choose what the assistant can access:
You can read any ImageKit documentation page as plain-text Markdown by adding .md to its URL. For example, the plain-text version of this page is at https://imagekit.io/docs/build-with-ai.md. You can also use the Copy page option at the top of any page.
This helps AI tools consume our content, and it lets you paste a whole doc into an LLM. Plain text is better than scraped HTML because it contains fewer formatting tokens, includes content hidden in the default view (such as text inside a tab or accordion), and preserves a Markdown hierarchy that LLMs can parse.
We also host an /llms.txt file, an emerging standard that indexes every page and links to its plain-text version. For everything in one request, /llms-full.txt concatenates the entire documentation into a single file. Point any AI agent at these so it answers from the latest ImageKit documentation instead of stale training data.
If an MCP client fails to connect:
For more help, contact ImageKit support.