Write Markdown Once, Publish It Everywhere: dev.to, Medium, AWS Builder Center and LinkedIn
This article provides a step by step guide to publishing one markdown article to dev.to, Medium, AWS Builder Center and LinkedIn with publishing-kit, an agent skill for Claude Code, Codex and Antigravity. The article you are reading was published with it, and every command output below comes from that run. https://github.com/xbill9/publishing-kit One markdown file goes in. A dev.to draft, a…
This article provides a step-by-step guide to publishing one markdown article to four different platforms: dev.to, Medium, AWS Builder Center, and LinkedIn. The article was published using a tool called publishing-kit, an agent skill for Claude Code, Codex, and Antigravity. Publishing the same markdown file to these platforms results in different outputs, as each platform treats the file differently. This guide aims to address these differences and ensure consistent publishing across all platforms.
Markdown is relatively easy to work with, but getting it published on four different platforms requires different handling of certain elements. For example, dev.to supports full REST API for creating, updating, and listing content, while Medium and LinkedIn do not have publishing APIs. The publishing-kit tool handles these differences by packaging workarounds as an agent skill and small Python scripts.
The guide walks through the entire process of writing a source article in markdown format with front matter, making a cover image, and using the publishing-kit to publish the same article to the four platforms. The process involves writing the source article, creating a cover image with the correct specifications for each platform, running the Python scripts for each platform, and checking the pre-flight conditions before publishing.
The publish-devto.py script is used for dev.to, make-medium.py for Medium, make-builder.py for AWS Builder Center, and make-linkedin.py for LinkedIn.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.