Projects

April 2026

Friday for Codex

macOS automation, voice activation, LaunchAgents, custom XTTS voice config, OpenAI Codex context workflows, Agent Driven Development

Say "wake up daddy's home" and Codex opens with a spoken, Friday-voiced briefing on what to work on next. Iron Man energy, on macOS.

Visit source
Friday for Codex cover image

Reality on a budget can't usually compete with science fiction backed by millions of dollars in CGI, but I got close enough to make my inner child happy.

Friday for Codex is a wake briefing and voice workflow for OpenAI's Codex on macOS, inspired by Friday and Jarvis from Iron Man. There's no big "future of work" thesis here. I wanted Codex to wake up like Tony Stark's lab assistant, and now it does.

The wake flow

You say the wake phrase. Codex opens, a greeting plays almost instantly, and then it reads you a briefing on what to pick up next, based on your recent work.

The default wake phrase is exactly what you think it is.

"wake up daddy's home"

Ridiculous, yes. Also the correct line.

Step by step, it:

  • listens for the wake phrase
  • opens Codex on macOS
  • plays a greeting right away
  • reads out a cached briefing
  • refreshes the briefing in the background so the next wake is fast

You never wait for an agent to think. The greeting plays right away, and the slow context work is already cached.

Under the hood

Everything runs locally. The briefing pulls from the places I check when I'm getting back into work:

  • Codex context
  • Codex Chronicle summaries, if available
  • recent ChatGPT conversations
  • Gmail
  • Google Calendar

The voice comes from a local XTTS setup with a custom-tuned voice model, as close to Iron Man's Friday as I could get it. A background job refreshes the cached briefing every 30 minutes. Without the cache, every wake would stall while the system gathers context and generates audio. With it, you call, you hear the briefing, and you get back to work.

Why bother

It isn't a general-purpose assistant platform. It's personal, it's local, and it's built around how I work. It reads my recent tasks, conversations, calendar and email, and speaks one briefing on top of the Codex workflow I already had.

You can watch the demo along with the Iron Man 2 scene that made me want to build this. The scene explains the motivation better than I can.

Not every project has to be profound. Sometimes the point is that your computer finally does the thing your 10-year-old self thought computers should obviously do.

Projects

April 2026

Friday for Codex

macOS automation, voice activation, LaunchAgents, custom XTTS voice config, OpenAI Codex context workflows, Agent Driven Development

Say "wake up daddy's home" and Codex opens with a spoken, Friday-voiced briefing on what to work on next. Iron Man energy, on macOS.

Visit source
Friday for Codex cover image

Reality on a budget can't usually compete with science fiction backed by millions of dollars in CGI, but I got close enough to make my inner child happy.

Friday for Codex is a wake briefing and voice workflow for OpenAI's Codex on macOS, inspired by Friday and Jarvis from Iron Man. There's no big "future of work" thesis here. I wanted Codex to wake up like Tony Stark's lab assistant, and now it does.

The wake flow

You say the wake phrase. Codex opens, a greeting plays almost instantly, and then it reads you a briefing on what to pick up next, based on your recent work.

The default wake phrase is exactly what you think it is.

"wake up daddy's home"

Ridiculous, yes. Also the correct line.

Step by step, it:

  • listens for the wake phrase
  • opens Codex on macOS
  • plays a greeting right away
  • reads out a cached briefing
  • refreshes the briefing in the background so the next wake is fast

You never wait for an agent to think. The greeting plays right away, and the slow context work is already cached.

Under the hood

Everything runs locally. The briefing pulls from the places I check when I'm getting back into work:

  • Codex context
  • Codex Chronicle summaries, if available
  • recent ChatGPT conversations
  • Gmail
  • Google Calendar

The voice comes from a local XTTS setup with a custom-tuned voice model, as close to Iron Man's Friday as I could get it. A background job refreshes the cached briefing every 30 minutes. Without the cache, every wake would stall while the system gathers context and generates audio. With it, you call, you hear the briefing, and you get back to work.

Why bother

It isn't a general-purpose assistant platform. It's personal, it's local, and it's built around how I work. It reads my recent tasks, conversations, calendar and email, and speaks one briefing on top of the Codex workflow I already had.

You can watch the demo along with the Iron Man 2 scene that made me want to build this. The scene explains the motivation better than I can.

Not every project has to be profound. Sometimes the point is that your computer finally does the thing your 10-year-old self thought computers should obviously do.