Compare / Tokensitter vs Caveman
Tokensitter vs Caveman
Tokensitter vs Caveman
More than short replies. Compare the whole workflow.
Both address token waste around coding agents. Compare the current toolkit and its controls, not just the original terse-reply skill.
Which levers do you want active, and how will you check them?
Caveman's current README describes six parts, each with its own on/off switch: shorter replies, reading less, automatic model and effort selection, reusable script blocks, waste fixes and compressed browsing. A skill-only install is also documented.
| Compare | Tokensitter | Caveman |
|---|---|---|
| Reply style | Chat reduction has Off, Lite and Ultra; Ultra is the default for users who never set it. | Reply styles are invoked as /caveman, /ultracave and /megacave. [1] |
| Beyond replies | Read less hooks, a lean-code instruction, a cheaper helper for read-only work, and separate relay levers. | Documents shrinking logs and test output, choosing model and effort per request (Auto, needs a free account), reusable script blocks (marked pre-release), waste fixes and compressed browsing. [1] |
| Controls | Widget controls with a detail view for each lever. | Per-part switches with caveman on and off, plus caveman status. [1] |
| Evidence | Lab results are on the Benchmarks page. | Links its own benchmarks and outside evaluations. [1] |
Competitor descriptions follow the linked official pages, not our own testing. A capability not mentioned here is not a claim that it is absent.
Choose for your workflow
Consider Tokensitter when you want a plan-and-app overview around several levers. Consider Caveman when its reply styles and toolkit match your agent workflow. They may both add instructions to the same session, so test before enabling both; we have not measured them together.
We have no controlled head-to-head benchmark against Caveman, so this page makes no claim about which saves more or answers better. Our own lab results, with their limits, are on Benchmarks. These are fixed tasks run in a test harness, not your sessions. Your saving depends on your prompts, your files and how long your sessions run.
Sources
- Caveman repository and README (https://github.com/JuliusBrussee/caveman), accessed 27 October 2026
Pages change. Check the linked sources before you choose a setup.