Skip to content

menubar: Run headroom proxy to save input + output tokens #597

Description

@reneleonhardt

https://github.com/headroomlabs-ai/headroom is an amazing tool to reduce input and output token usage.
But it only works natively for CLI apps, for desktop apps like Codex you need to manually run headroom proxy --port 8787.

Would it be possible to integrate headroom into the menubar (start if not already running), so the user doesn't forget and need an open terminal before opening the desktop assistant?

# Install
uv python install 3.14
uv tool install "headroom-ai[all]"

# .zshenv / .bash_profile
export HEADROOM_EMBEDDER_RUNTIME=pytorch_mps    # macOS only
export HEADROOM_OUTPUT_SHAPER=1
export ANTHROPIC_BASE_URL=http://127.0.0.1:8787 # Claude Code
export OPENAI_BASE_URL=http://127.0.0.1:8787/v1 # Codex / OpenAI

# Run proxy before desktop app (blocking)
headroom proxy --port 8787

# Open desktop app
open -a Claude
open -a Codex

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Fields

    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions