Developer tools decision
Headroom
A competent developer can reproduce Headroom's core (local proxy + reversible compression + per-project learnings) in a few weeks; prior art (rtk) shows the core is achievable, so building is realistic and cheaper than continued paid use for a single user.
Visit website↗Not priced
No pricing page we fetched carried a figure, so there is nothing to compare against. The build side is still real.
$100one-off60 h to build
$0/mo3 h/mo upkeep
No published price to break even against.
Open-source builds that already do this
Every project below is open source and already does this job today. Fork one, self-host it, or take the parts you need - the build prompt further down assumes an empty file, and this is the shortcut past that. Licences differ; check the one on each card before you ship. All Headroom alternatives, with the arithmetic →
What a replacement has to do
- Run a local proxy that intercepts prompts/tool outputs, compress noisy JSON/log/HTML blobs reversibly, forward compressed prompt to the LLM client and restore original content on demand, persist per-project token-saving learnings, and expose minimal CLI telemetry.
What it still won’t have
- Signed & notarized macOS installer and automatic updates (desktop app polish)
- Menu-bar native UI and first-party desktop UX
- Commercial support / priority support
- The paid app's packaging (one-click installer) and paid rollout controls
What remains hard
- Product polish and ongoing maintenance
First-year cost
No published price
Headroom does not publish a price we could read, so there is nothing to compare against. What building costs is below.
Money you would actually spend
Time you would spend
—
What you would spend
What we assumed
The verdict above measures whether you could build it. This one is only about money.
Runnable build prompt
Build a cross-platform local CLI proxy in Rust (single binary) that intercepts HTTP(S) requests from Claude Code/Codex clients and performs reversible token-aware compression of noisy inputs (JSON, logs, HTML). Core features in scope: 1) request interception and forwarding, 2) compressor/decompressor module that preserves retrievability, 3) per-project storage for learnings (text files) and an activity log, 4) a CLI to start/stop proxy, toggle add-ons (RTK-like trimming, MarkItDown conversion), and show daily token-savings summary, 5) tests for compression correctness and end-to-end proxying, and robust error handling and logging. Out of scope: signed macOS installer, GUI menu-bar app, hosted telemetry, and paid-support tooling. Deliverables must include unit tests, an end-to-end integration test that runs a mock Claude client through the proxy, and documentation for local installation and usage.
How we checked
How the score was reached
- Partly verdict base52
- An open-source build was found+5
- 4 cited sources+3
- Evidence score60
The base comes from the verdict. Everything under it is a check that either happened or did not, and each one is a fact frozen in this record rather than a judgement made at render time - so the same evidence always produces the same number.
How scoring works →Cited sources · 4
Every page the run actually retrieved.
- official productHeadroom — Cut Claude Code & Codex Costs ~50%
- official docsHeadroom Features: Add-ons, Project Learnings & Activity
- official pricingClaude Code Usage: Stretch Your Plan 2x | Headroom
- open sourcertk-ai/rtk
Integrity checks
What held up, and what did not.
