Share images · Illustrations · Infographics

Image generator skills for AI agents

Art direction your agent can repeat: a written brief, a fixed visual system and a quality check, rendered with PopNow image models at a price you approve first.

Works inClaude CodeCodexCursorGemini CLIGitHub CopilotOpenCode

What do you want to make?

EverythingThumbnails & coversInfographicOG imageArticle illustrationsStoryboardTechnical posterIllustrated map

One look, every time

Each skill fixes the palette, composition and type rules, so a series of images reads as one set.

Exact words

Headlines are checked letter by letter, and text-heavy layouts use models that render type cleanly.

Finish in place

Upscale, cut out a transparent PNG or make a variation without leaving your agent.

How agent skills work

  1. Connect or install once

    Connect PopNow to Claude or ChatGPT, or install the skill in your coding agent.

  2. Ask in plain words

    Call the skill and describe the job. It follows its own playbook, layout rules and checks.

  3. Approve any spend

    When a step needs AI images, video or sound, it quotes PopNow credits and waits for your yes.

Questions

What is an agent skill?

An agent skill is a folder with a SKILL.md playbook plus templates, scripts and references that an AI agent follows when the task matches. Claude Code, Codex, Cursor and other agents load it automatically, so the result follows a tested design system instead of a one-off prompt.

What is the difference between agent skills and MCP?

MCP connects an agent to tools and data, such as PopNow's image and video models. A skill is a playbook that tells the agent how to do a whole job well: the layout rules, the steps and the checks. The two work together: connect PopNow over MCP and your assistant can find these skills and use them, and the skills call PopNow for pictures and sound.

Are the skills free?

Yes, every skill is free with a PopNow account, in Claude or ChatGPT through PopNow or installed in a coding agent. Only the images, video, music or voice a skill makes use PopNow credits, and each one is priced and approved before it runs.

Can I use these skills in Claude or ChatGPT?

Yes. Connect PopNow to Claude, ChatGPT or another app that supports MCP, then say what you want to make. Your assistant shows you matching styles, uses the one you pick and checks the price with you before it generates anything. Skills that build files, such as decks, websites and HTML video, work best in a coding agent like Claude Code.

Which agents are supported?

Any agent that reads SKILL.md folders: Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot and OpenCode. The skill is the same everywhere; only the install folder differs.

How is this different from PopNow's own skills?

Skills on the main Skills page run inside PopNow with Poppie. Agent skills run in your own coding agent on your computer, which suits files you want to keep editing, such as decks, websites and HTML-based video.

Can I use the results commercially?

Yes. What you make is yours, including client work. The skill package itself is for your own use and should not be resold or redistributed.

Which AI tool makes product launch videos?

PopNow does: connect it to Claude or ChatGPT, or install a launch video skill in a coding agent, and pick one of more than 20 free launch video styles. The agent renders the MP4 from your real product and adds PopNow music, voice and sound effects after you approve the price; see How to make a product launch video with AI.

What is the best way to make video, images and decks with an AI agent?

Give the agent a skill for the job and a place to make the media: PopNow agent skills are written for PopNow, so the layouts, prompts and model choices land on the look first time, and the pictures, clips, music and voice come from one PopNow account with every run priced up front.