Back to Projects

In Development

Browser Extension

iOS Shortcut

AI Brain

OmniInbox

Zero-friction capture from voice, screen, and research — straight into action.

A multi-surface intake system that turns voice memos, browser captures, and AI research exports into classified, actionable items routed to the right destination automatically.

Story

Most capture tools are single-channel — a voice recorder, a bookmarking extension, a note-taking app. OmniInbox unifies all of them behind one pipeline: audio from an iPhone shortcut hits a FastAPI daemon over Tailscale, a Chrome extension grabs screenshots and full Gemini Deep Research articles, and everything flows through Whisper transcription, Claude classification, and skill-based action routing. The result is a personal operations layer where nothing captured is ever lost or left unprocessed.

Focus

Multi-channel AI-powered capture, classification, and action routing for personal knowledge operations.

Product Type: Agent System

Technical Highlights

  • Voice capture via iPhone Shortcut to FastAPI daemon over Tailscale — speak and forget.
  • Chrome extension captures screenshots and extracts full Gemini Deep Research articles as clean markdown.
  • Whisper transcription with Claude-powered multi-intent classification and automatic action routing.
  • Built-in skills: GitHub issue creation, reminders, project logs, email summaries, and LinkedIn drafts.
  • Verbatim pipeline exit for finished content — deep research articles saved to Obsidian with zero summarization loss.

How to provide media for this page

Add screenshots and demo clips into public/media. Then register assets in data/projects.ts and reference them in this template. For YouTube, include a public link and we can embed it directly.