- Home
- Skills
- Image & Video
- Code Card
Works with the AI Agents you already use
Code Card
Render code into Carbon-style high-res code card PNGs; single snippet or Markdown batch; offline-capable with embedded highlight.js.
Free
Code Card
Example session with this skill installed
Convert the code blocks in this README.md to images using the nord theme.
- Read your context and instructions
- Compiled the code card
I have processed README.md and converted the detected code blocks into Carbon-style images using the Nord theme.
The original code blocks have been replaced with image references in a new file, and ASCII box-art diagrams were preserved as text.
Render Summary
| Type | Count | Action |
|---|---|---|
| Code Blocks | 3 | Rendered to PNG |
| ASCII Art | 1 | Preserved as text |
| Output File | 1 | README-cards.md |
| Image Path | - | ./README-code-images/ |
Generated References in README-cards.md
# Project Overview
Check out our main logic:

And the configuration setup:

Next steps
- Review
README-cards.mdto ensure the image placement is correct. - Run
mv README-cards.md README.mdif you want to overwrite the original file. - Upload the
./README-code-images/folder alongside your Markdown file when publishing.
Connects securely to your tools. The creator never sees your data.
About this skill
Sharing raw code blocks in documentation or articles often looks unpolished and lacks visual impact. This skill converts plain text code into high-definition, Carbon-style images with macOS window decorations and syntax highlighting.
What it does
- Carbon-style rendering: Generates high-definition (3x Retina) PNG images featuring macOS window controls and line numbers.
- Batch processing: Parses Markdown files to detect code blocks, renders them as images, and replaces the text blocks with image references automatically.
- Language auto-detection: Uses
highlight.jsto infer languages from file extensions or Markdown tags for accurate syntax coloring. - Idempotent output: Uses content hashing for filenames to prevent duplicate image generation on repeated runs.
- ASCII preservation: Intelligently skips ASCII box-art diagrams to keep architectural drawings as editable text.
- Theme support: Includes 6 built-in themes like Dracula and Nord, plus support for custom JSON theme definitions.
How it works
- Identify: Point the agent to a
.mdfile for batch conversion or any source code file for a single snippet render. - Render: The skill uses a local
playwrightinstance to capture the code in a headless browser environment. - Replace: For Markdown files, it generates a new
-cards.mdfile with image links while preserving the original structure. - Finalize: Images are saved to a local directory for deployment or manual sharing.
Frameworks & tools
Built with Node.js and Playwright. It uses a local copy of highlight.js for offline syntax highlighting and supports system-installed Chrome/Edge to avoid heavy binary downloads.
Why this beats prompting it yourself
Generic LLMs cannot generate actual PNG files or handle the layout logic of window borders and retina scaling. This tool automates the multi-step process of rendering, file management, and Markdown reference updating that would otherwise require manual screenshotting and editing.
Use cases
- Technical blogging: Convert README code snippets into social-media-ready images for platforms like WeChat or X.
- Documentation: Standardize the look of code examples across a large Markdown-based documentation site.
- Social sharing: Quickly turn a local Python or Rust file into a polished "Carbon" image for sharing.
- Presentation slides: Generate consistent, high-contrast code visuals for talks and demos.
Known limitations
Requires Node.js and the Playwright npm package in the execution environment. High-quality Chinese font rendering depends on the host system having those fonts installed.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Video Skills.
Requires Node.js and Playwright. Best results on systems with Chrome/Edge installed. Chinese characters require system fonts. Skip logic for Mermaid/PlantUML blocks.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 12 days ago
- Free with an Agensi account
- Software
- Playwright
Node.js