Give Each Request a Clear Route
A focused Skill tells the agent when to use it, which inputs it accepts, what process to follow, and what result to return.
EZREMOVE FOR AGENTS
Install task-specific EzRemove Skills once, then ask your agent for a clear outcome without rebuilding the process in every conversation.
Install EzRemove Skill for me: https://github.com/ezremove/skills.gitWorks with the agents you already use
✓Free, unlimited Remove Background API with a Skill API key
AVAILABLE SKILLS
Each EzRemove Skill is built around one defined job, with documented instructions, accepted inputs, and a predictable result. Install only the capabilities that fit your workflow and return to it as new Skills become available.
The first available EzRemove Skill turns a local image or image URL into a clean, transparent PNG. It is ready to install now for product photos, graphics, profile images, and reusable cutouts.
ONE-LINE INSTALL
Copy one instruction into a compatible agent. The agent reads the EzRemove Skills package, installs the selected capability, and guides you through any required setup. You do not need to rebuild instructions or assemble the workflow manually.
Install EzRemove Skill for me: https://github.com/ezremove/skills.gitNATURAL-LANGUAGE WORKFLOW
Keep the request, source material, and returned result together in one conversation. Describe what you need in plain language, and your agent uses the matching EzRemove Skill to follow a defined process.
WHY EZREMOVE SKILLS
A focused Skill tells the agent when to use it, which inputs it accepts, what process to follow, and what result to return.
EzRemove Skills are structured around assets and results that can continue into the next part of a real workflow, rather than ending with a generic chat response.
The package is hosted on GitHub with readable instructions, scripts, and references, so you can understand a Skill before asking your agent to install it.
Connect EzRemove Skills with your AI agent using a Skill API key. Process tasks without per-image charges or usage limits through a simple workflow.
HOW IT WORKS
Open the available Skills list, review the capability and expected result, and select the Skill that matches the task you want your agent to perform.
Paste the one-line instruction into Codex, Claude Code, Cursor, or OpenClaw. Your agent reads the package and prepares the selected Skill.
Provide the relevant input and describe the outcome in plain language. Your agent follows the installed Skill and returns the result in the conversation.
COMMON QUESTIONS
An AI agent Skill is a reusable set of instructions and supporting resources that helps an agent complete a defined task with a more consistent process and result.
EzRemove Skills are focused capabilities that extend what a compatible agent can do. Each Skill documents its purpose, inputs, workflow, and expected output.
The current EzRemove experience is designed for Codex, Claude Code, Cursor, and OpenClaw.
Remove Background is the first available Skill. It accepts a local image or image URL and returns a transparent PNG. Additional Skills can be added to this homepage as they launch.
No. Send the install prompt to your agent and describe the result you need in ordinary language. The agent reads the Skill instructions and guides you through required setup.
Yes. The EzRemove Skills package is hosted on GitHub, so you can review its instructions, scripts, and references before installation.
No. This homepage is the expandable entry point for EzRemove Skills across task categories. Remove Background is simply the first available capability, not the limit of the library.
Use the install prompt with https://github.com/ezremove/skills.git in a compatible agent. The agent handles the setup steps from there.