Show HN: Claude Code's $200 plan is a 17× subsidy on the raw API (github.com)

A developer recently introduced a unique analytical tool for users of Claude Code, detailing how costs are incurred during API usage. By reverse-engineering logs from their own usage, they uncovered that the majority of billing is linked to re-reading context, with about 64% of total expenses coming from regenerating previously sent entire contexts. In a month of use, this individual spent approximately $3,371 at Opus 4.7 rates, with input re-reading dominating the costs. The tool allows users to break down their own usage into insights on token usage, time spent, and expense distribution, showcasing that caching significantly mitigates costs. This revelation is significant for the AI/ML community as it highlights the importance of context management in optimizing API expenses, especially for developers relying on Claude Code for complex interactions. The method encourages users to analyze and understand their own usage patterns, promoting efficiency and potentially saving on costs. With the ability to run the analysis personally, users can tailor insights specific to their interaction patterns, empowering developers to make informed decisions and refine their strategies for API usage.