Tokendoctor cuts Claude Code costs by up to 38% with settings it can undo

Tokendoctor – the same Claude Code for up to 38% less

Tokendoctor cuts Claude Code costs by up to 38% with settings it can undo

A Claude Code plugin that replays your own sessions at smaller context windows to show what you would have saved, then audits 38 settings against current docs and applies the fixes you approve as diffs. One trick, called the slide, swaps old tool outputs for summaries where Claude Code would otherwise compact the conversation. Everything runs locally, and any change can be reverted with no model call. Measured on its author's six weeks of use: 38% lower main-thread cost.

There is no one pill. If your sessions rarely pass 400K tokens, the window changes nothing for you; a summary that lands just before you finish can cost more than it saves; your models, effort and work all move the number.
  1. urig

    Please don't let AI write your README.md. It's illegible to humans.

  2. Incipient

    All of these token optimiser things...i rarely see any performance analysis. If you're cutting token usage in half, what does that do to the accuracy of the outcome? If you're automatically delegating to cheaper models, what does that do to the outcome?

    Personally I've found a HUGE improvement in quality using more context, to the point that it's 'better value' - fewer outcomes I disagree with or have to fix/rework.

    I've mostly done this through mechanical documentation links that are enforced and can be 'followed' with some python files that the agents use. Context is higher than just reading code, but it has been a HUGE benefit (as far as I can tell).

  3. relyks

    I use this and it works well: https://github.com/rtk-ai/rtk

    I'm trying to figure out if I can use multiple of these tools together without conflicts

More from this day

2026-11-17