Headroom is an open-source context-compression layer for AI agents that shrinks tool outputs, logs, JSON, files, RAG chunks and conversation history before they reach the LLM, running entirely on your machine.
It can be used as a Python or TypeScript library, a zero-code proxy, an MCP server, or by wrapping existing coding agents such as Claude Code, Codex, Cursor and Aider with one command.
- 01
`headroom wrap <agent>` routes Claude Code, Codex, Copilot, Cursor, Aider, Cline, Goose and others through Headroom
- 02
Zero-code proxy (`headroom proxy`) for any language plus Python/TypeScript `compress()` library
- 03
Content-aware compressors for JSON (SmartCrusher), code (AST-based) and text
- 04
Reversible compression: originals cached locally and retrievable on demand
- 05
MCP server with headroom_compress, headroom_retrieve and headroom_stats tools
- 06
Cross-agent memory and `headroom learn`, which writes corrections into CLAUDE.md or AGENTS.md
Open Source
Headroom OSS is free and open source under Apache-2.0 and runs locally. Headroom Enterprise adds managed or in-VPC rollout, SSO, central config and dedicated support at custom pricing; companies signing a 12-month Enterprise plan get the first 3 months free.
Best for
Developers and teams running coding agents on large logs, JSON or codebases who want to cut token spend and context pressure without changing their agent or code.
Not ideal for
Workflows that need exact verbatim prompts to the model at all times, or teams unwilling to run a local proxy in front of their LLM provider.
A widely adopted, actively maintained way to make coding agents cheaper and less context-starved, with unusually broad agent integrations and a local-first design.
v0.40.0 (2026-10-06) made proxy usage reporting opt-in and added HEADROOM_LICENSE support, per-message compression diagnostics, a lossless live-agent densify mode, Antigravity CLI as a learn backend and a per-request keep-last-turns header.
As of
Is Headroom free?+
Yes - Headroom is Open Source. Headroom OSS is free and open source under Apache-2.0 and runs locally. Headroom Enterprise adds managed or in-VPC rollout, SSO, central config and dedicated support at custom pricing; companies signing a 12-month Enterprise plan get the first 3 months free.
Is Headroom open source?+
Yes - Headroom is open source. Headroom OSS is free and open source under Apache-2.0 and runs locally. Headroom Enterprise adds managed or in-VPC rollout, SSO, central config and dedicated support at custom pricing; companies signing a 12-month Enterprise plan get the first 3 months free.
Who is Headroom best for?+
Developers and teams running coding agents on large logs, JSON or codebases who want to cut token spend and context pressure without changing their agent or code.
Who is Headroom not ideal for?+
Workflows that need exact verbatim prompts to the model at all times, or teams unwilling to run a local proxy in front of their LLM provider.
What are the best Headroom alternatives?+
The closest Headroom alternatives on ai.dosa.dev are Pieces for Developers, agenttrace, Vibe Kanban - all listed under Developer Productivity & Workflow.
Who makes Headroom?+
Headroom is developed by Headroom Labs. It is listed in the Developer Productivity & Workflow category on ai.dosa.dev.
Favorite this tool to revisit it later, or Zap it to contribute to the public vote count.