Skip to main content

Halo Architecture and Runtime

1. The workspace shape​

The README reveals a system built from several distinct ideas:

AreaWhy it matters
workstation product layerthe main user-facing shell
pluggable agent enginesconnect Claude Code, Codex, and related backends
digital human subsystemruns scheduled autonomous work
browser actionsreusable deterministic scripts for web tasks
stores and distributionmake workflows reusable and organization-friendly

That is a large product footprint compared with a normal coding CLI.

2. The runtime mental model​

Halo generally works like this:

  1. the user or scheduler defines a task,
  2. an agent engine reasons about the task,
  3. browser actions or tools perform stable operations,
  4. outputs return into the workstation,
  5. remote-control surfaces can monitor or steer the process.

3. Why browser actions matter​

Halo explicitly argues that AI should decide what to do while prewritten browser actions handle how to do it. That design matters because it reduces the brittleness of pure free-form browser agents.

4. Digital humans are a first-class subsystem​

The project treats autonomous scheduled workers as a real product category, not as a thin cron wrapper. That is one of the clearest differences between Halo and ordinary desktop AI apps.

5. What to read first in code​

Start with:

  1. the README product overview,
  2. browser-action and digital-human docs,
  3. desktop and runtime structure,
  4. integration and store surfaces.