Headroom open-sources a local-first way to cut agent token use by up to 95%
eyishazyer · x · 2026-07-23
Headroom compresses agent context to cut token use by 60–95%
An ex-Netflix engineer has open-sourced Headroom under Apache 2.0, a local-first repository that compresses everything an agent reads before sending it to the model.
- It targets tool outputs, long logs, RAG chunks, files, code, and conversation history.
- The author claims it can reduce JSON token usage by 60–95% and coding-agent tokens by about 15–20%.
- It keeps original files on your machine and lets the agent fetch them again when needed.
- It can plug into Claude Code, Codex, Cursor, Aider, and Copilot via a library, proxy, or MCP server.
- Reported benchmarks say accuracy was preserved on tasks like GSM8K and TruthfulQA.
More from coding & agent
- Ben Bajarin: Frontier AI is becoming the enterprise cyber-defense stack — BenBajarin · 2026-10-26
- Building AI bots on raw APIs? Devs say models are 'dumb without the harness' — madeofAjala · 2026-10-26
- HarnessMM cuts agent tokens per task by up to 88% while raising success rates — omarsar0 · 2026-10-26
- Microsoft demos intent-classifying agents built with Copilot App and Foundry, deployable across Teams and Outlook — marlene_zw · 2026-10-26
- Looking for a wearable mic with best battery life that feeds recordings to personal agents via MCP — altryne · 2026-10-26
- The identity crisis in software engineering: multiagent systems demand a new problem set — rakyll · 2026-10-26