Hiten Shah Defines Re-Explanation Tax And Launches Free Claude Plugins 101
Reactions from ranked influencers
2 posts
Every time you have to explain the work to AI again, you pay a tax. You explain who the work is for and what good looks like. Claude needs another reminder about how the reasoning should be structured. Then you repeat the correction you made two days ago. None of this feels expensive in the moment. Across a company, it becomes lost time and uneven work. Important checks disappear while old information persists. Useful judgment remains trapped in the heads of the people with the most experience. The result depends on who happened to configure Claude that day. I call it the re-explanation tax. Skills can preserve a method you keep teaching Claude. Plugins can carry more of the workflow, including the tools and specialist capabilities the work depends on. The company becomes less dependent on how well one person happens to prompt Claude. It can begin preserving and distributing what its best people already know. Tomorrow at 10 AM PT, I’m teaching Claude Plugins 101 live. We’ll make three decisions painfully clear. When is a prompt enough? When should a repeated correction become a skill? When does the broader workflow deserve a plugin? Free live session: http://hiten.com/plugins-101
The best thing you taught Claude this week is probably trapped in a chat you’ll never reopen. A correction becomes valuable when it survives the conversation. The instruction for today may belong in a prompt. The method you keep repeating may deserve a skill. The broader workflow may belong in a plugin. At 10 AM PT, I’m teaching a simple framework for making that decision. By 11, you’ll know where your reusable AI work belongs. Free live sessions http://hiten.com/plugins-101