# Headroom

Hacker Approved page: https://hackerapproved.com/p/headroom/

9 people recommended Headroom in Hacker News comments, 4 had mixed feelings and 5 criticized it, November 2025 to October 2026. It was recommended in 4 of 12 months, in 7 threads. It does not meet the Hacker Approved badge rule.

**Verdict:** Free open-source layer that compresses tool outputs before they reach the LLM, cutting token use. Critics say savings are unproven and it can break prompt caching.

Summary written by Claude from 30 comments. It was written before the November 2026 recount, and 24 of those comments still count.

## Why people recommend it

- Cuts token use and cost (3 comments in the sample): "tools like RTK, Headroom, caveman mode and others do reduce input and output tokens that need to be processed" ([source](https://news.ycombinator.com/item?id=48592020)) "setting up a combination of both tools could reduce token consumption by 50%" ([source](https://news.ycombinator.com/item?id=48595013)) "the real tool you want to reduce token usage is headroom" ([source](https://news.ycombinator.com/item?id=47795229))
- Transparent benchmarks and algorithms (3 comments in the sample): "They provide accuracy benchmarks in the repo and are transparent about the compression algorithms used for the different output types." ([source](https://news.ycombinator.com/item?id=48593858)) "I like headroom’s approch a lot better. More transparent." ([source](https://news.ycombinator.com/item?id=48593864)) "I applaud the benchmarking though." ([source](https://news.ycombinator.com/item?id=48601689))
- Keeps needed data, reversible (2 comments in the sample): "compresses tool outputs (schema-preserving subset of JSON arrays; keeps errors/anomalies/top/relevant" ([source](https://news.ycombinator.com/item?id=46602762)) "The goal here was: reduce tokens without breaking the contract." ([source](https://news.ycombinator.com/item?id=46602779))
- Works with Claude Code and Cursor (1 comment in the sample): "you've made it a joy to use!  I've just made the v0.0.1 of an adapter to run headroom as a hook" ([source](https://news.ycombinator.com/item?id=48500657))

## What people warn about

- Token savings unproven or negligible (4 comments in the sample): "with some limited "success" (for some definition of success" ([source](https://news.ycombinator.com/item?id=49700858)) "Benchmarks done on Headroom and RTK show that neither result in real savings." ([source](https://news.ycombinator.com/item?id=49657411)) "the rest of the efforts didn’t move the needle much" ([source](https://news.ycombinator.com/item?id=47740454))
- Can raise conversation cost (2 comments in the sample): "the number of turns to complete went up so the overall cost of the conversation was higher" ([source](https://news.ycombinator.com/item?id=48174550)) "headroom violates cache frequently, nullifying any potential savings it could have" ([source](https://news.ycombinator.com/item?id=49665630))
- Hard to configure (1 comment in the sample): "That was a real PITA to configure for local-only" ([source](https://news.ycombinator.com/item?id=49795367))
- Lacks transparency about how it works (1 comment in the sample): "The lack of transparency here is really concerning." ([source](https://news.ycombinator.com/item?id=47803166))

## The badge

A product earns the Hacker Approved badge when at least 10 people recommend it, recommenders outnumber critics at least 2 to 1 (mixed feelings count toward neither side), it was recommended in at least 4 of the 12 months and in at least 5 different threads, and no single thread supplies more than half of the people who recommend it. Health and money products are never badged.

## Cite as

Hacker Approved, "Headroom", https://hackerapproved.com/p/headroom/ (9 recommend, 4 mixed, 5 criticize; Hacker News comments, November 2025 to October 2026).

Counts updated 2 Nov 2026, with data through October 31, 2026.