# headroomlabs-ai/headroom

> A local library, proxy and MCP server that compress tool outputs, logs, files and RAG chunks before they reach the LLM, cutting token usage without changing answers.

- Magnitude: 3.2 out of 10 — Steady
- Stars: 74,807 total · +7 stars measured 2026-10-09, 11:36–13:11 UTC, ≈ 39 by evening
- Star trust: star growth looks organic
- Category: Language models · Language: Python · License: Apache-2.0 · Created: 2026-01-07 · Last push: 2026-10-09
- GitHub: https://github.com/headroomlabs-ai/headroom · Homepage: https://docs.headroomlabs.ai/docs · Page: https://gitnova.dev/en/r/headroomlabs-ai/headroom

## Useful for

- Wrap Claude Code or Cursor with headroom wrap to cut token usage in a session
- Run headroom proxy and route any app's requests through it without code changes
- Compress large JSON outputs from MCP servers via compress() in Python or TypeScript

## Why it’s here

- Star-counter measurements on 2026-10-09 (UTC), 11:36–13:11: 74800 → 74807 stars (+7). This is the change over that interval.
- Estimated end-of-day forecast: about +39 stars, using observed gains and the previous day.
- GitHub Trending Python today: #7, +120 stars.
- About 26 forks a day — people are taking the code.

## Star trust

Star growth looks organic.

Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer.

## Numbers

- Forks: 5,801
- Issues and pull requests: 3,927
- Watchers: 217
- Average over the last week: 84 per day
- Usual pace: 134 per day
- Stars in the last hour (measured): 4
- Latest release: v0.40.0 (2026-10-06)

## Stars per day, last 30 days (oldest → newest, today is partial)

2026-09-10 … 2026-10-09: 287, 206, 160, 190, 250, 158, 196, 282, 169, 166, 181, 137, 140, 123, 88, 91, 85, 83, 86, 95, 82, 72, 65, 70, 73, 53, 82, 117, 150, 7

## Spotted in now

- GitHub Trending Python today: #7, +120 stars

## Similar by description

1. **JuliusBrussee/caveman** — 3.4 · Steady · AI agents · Go · +103 stars measured 2026-10-09, 00:07–13:11 UTC, ≈ 188 by evening
   A skill and local proxy for coding agents that cut token usage: the skill makes the agent answer in terse caveman style, the proxy compresses logs, diffs and test output before they reach the model.
   Full card: https://gitnova.dev/en/r/JuliusBrussee/caveman.md
2. **rtk-ai/rtk** — 2.6 · Steady · AI agents · Rust · +33 stars measured 2026-10-09, 00:08–13:13 UTC, ≈ 66 by evening
   A Rust CLI proxy that filters and compresses bash command output before it reaches the LLM context, cutting token usage by 60–90%. It hooks into Claude Code, Cursor, Gemini CLI and other agents.
   Full card: https://gitnova.dev/en/r/rtk-ai/rtk.md

---

Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score.
Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast.
Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself.
Data as of 2026-10-09 13:25 UTC, updated every 30 minutes.