20 SepFinished my ClaudeCut experiment. I wanted to see how much of a $20 Claude subscription actually goes into the work, and how much gets eaten by Claude Code itself. I cut it down to one terminal tool, a 36-token system prompt, and a capped context window. Startup dropped from…ClaudeAnthropicClaude CodeContext managementAgent harnesses
29 JulJetBrains MythBusters tested the Ponytail skill for Claude Code Before that, they analyzed Caveman. Its promised 65% token savings turned into 8.5%, with no cost reduction. Then came RTK, claiming up to 90% savings. In practice, some tasks became more expensive. The new review…ClaudeAnthropicJetBrainsClaude CodeAgent harnesses
