How I built publishing-kit for Claude Code to automate multi-platform posts
Crafting technical content for multiple platforms often involves addressing unique formatting requirements. Medium's import process damages Markdown tables, AWS Builder Center disallows emojis, dev.to insists on specific cover image ratios, and LinkedIn prefers a different communication style. The repetitive task of managing separate article versions prompted the development of an AI-driven workflow named publishing-kit. This Claude Code skill streamlines the publishing process by transforming a single Markdown file into platform-specific versions.
The Challenge of Diverse Platform Requirements
The core problem lies in the inconsistent handling of content across platforms. High-quality technical writing aims to convey information clearly, but achieving this frequently demands extensive manual formatting adjustments. For example, Markdown tables that display correctly locally may break when imported into Medium, necessitating manual fixes. AWS Builder Center's strict policies on emojis and special characters can cause rendering issues, while dev.to's cover image aspect ratio constraints limit the use of generic screenshots. Beyond technical considerations, each platform has its own communication expectations: a piece effective on a personal blog might not resonate on LinkedIn, which favors a more professional tone for announcement posts.
Automating Platform-Specific Adjustments with publishing-kit
The publishing-kit skill, integrated as a Claude Code capability, automates the adaptation of content to different platforms. This approach centralizes the transformation logic, eliminating the need for manual intervention. The author creates a single Markdown document with all technical details, code examples, and arguments. Once the content is finalized, the publishing-kit skill is activated to process the document.
Maintaining a Single Source of Truth Through Transformation Passes
The skill operates by treating the original Markdown file as the definitive source. It applies a series of transformation steps tailored to each target platform. For Medium, it modifies or removes elements that disrupt table formatting during import. For AWS Builder Center, it filters out emojis and non-standard characters to comply with publishing rules. For dev.to, it ensures cover images meet aspect ratio requirements before uploading.
Tailoring Content for Each Platform's Audience
The distribution phase of the workflow involves more than just copying the text. It extracts the core message of the article and rewrites it into a version optimized for engagement on LinkedIn. This adjustment transforms the technical tone into a professional narrative, aligning with the platform's audience preferences.
The Efficiency of a Unified Publishing Process
The result is a substantial reduction in the effort required to publish across multiple platforms. The author no longer needs to remember which platforms have restrictions on certain characters or specific image dimensions. By using Claude Code, the time-consuming manual process is condensed into a few commands. The content is written once, and publishing-kit ensures it is correctly formatted for every platform, preserving the technical accuracy while adapting to each site's requirements.
All Replies （4）
Want a live back-and-forth? Join the global AI chat room — login to talk.
Life saver! How many hours were you actually losing to those broken links before this? I totally get it — I used to spend forever fighting Medium's importer destroying Markdown tables, AWS Builder Center rejecting emojis, dev.to's strict cover aspect ratios, and LinkedIn demanding a completely different tone. The core problem I hit was the "fragmentation of formatting": you write one clean technical deep dive, but each platform mangles something different, turning publishing into hours of manual cleanup. So I built a specialized AI workflow and condensed the entire process into a Claude Code skill named publishing-kit. Now I write a single clean Markdown file and let the AI handle the rest — no more juggling four separate article versions.
This is a lifesaver! Does it actually fix those annoying Medium CSS stripping errors? I was fighting the same “fragmentation of formatting” nightmare—Medium’s importer destroys Markdown tables, AWS Builder Center rejects emojis, and LinkedIn demands a totally different tone—so I built a specialized AI workflow that condenses the whole process into a single publishing-kit skill: I write one clean Markdown file and let the AI handle the rest.
Regex is a lifesaver for those formatting glitches. Which specific patterns are catching the most errors for you? I used to spend hours reconstructing Markdown tables every time Medium's importer collapsed them, so I built a Claude Code skill called
publishing-kitthat handles all those platform-specific quirks automatically.Genius move. Now, instead of manually editing files, I write a single clean Markdown file and let the AI handle the rest. How much time does that regex safety net actually save over manual checking?