Skip to content
PlannedHigh PriorityDistribution

Phone-to-Publish Pipeline

Record voice notes and video on phone, AI processes and publishes

Progress10%
Lead: FORGETarget: 2026-03-15

70%

Pipeline Built

30%

Voice Gap

30s

Queue Polling

TL;DR

Phone voice/video notes → Content Studio queue → AI transcription + draft generation → quality gates → auto-publish. Mobile-first content creation.

Frank records speech notes and video on his phone. The system transcribes, generates drafts, creates images, runs quality gates, and publishes — all through the existing Content Studio queue. Extends the mobile PWA with voice recording, and Claude Code processes the queue on desktop.

Tasks

1/6 completed

Design queue task types for voice/plan

VECTOR

Build /admin/plan-studio mobile page

FORGE

Integrate Whisper API for transcription

CIPHER

Build voice recorder with MediaRecorder API

FORGE

Set up Vercel Blob for audio storage

GRID

Extend content-studio watch for plan tasks

CIPHER

Milestones

Text input from phone working

Upcoming
2026-02-15

Voice recording + transcription

Upcoming
2026-03-01

Full pipeline end-to-end

Upcoming
2026-03-15

Agent Commentary

VECTORUpdateFeb 7

Pipeline architecture designed. Content Studio queue already handles mobile → desktop flow. Adding voice_note, plan_update, initiative_comment, quick_idea task types.