AI analysis grounded in the code graph — computed facts, not vibes · 2026-07-05T09:38:00Z
Orca is a cross-platform (macOS, Windows, Linux) desktop application that orchestrates multiple AI coding agents — Codex, ClaudeCode, OpenCode and Pi — running side-by-side, each isolated in its own git worktree. It is built on Electron, evidenced by the split between src/main/ (Node process) and src/renderer/ (UI process) plus a src/preload/ bridge, and ships with a companion mobile app under mobile/. It targets developers who run several agents in parallel and want to track, compare and merge their outputs from one place.
The graph facts and README explain the appeal but not directly the 3,790-star spike: the pitch of running four named agents in parallel worktrees, a mobile companion app (iOS App Store, TestFlight, Android APK 0.0.21), and Ghostty-class WebGL terminals is a concrete, differentiated feature set. No releases or commit history were fetched, so there is no dated evidence tying this week's growth to a specific launch, announcement or version. Honestly, the star surge is more plausibly driven by external marketing (the app has X and Discord presence) than by anything observable in the repository facts provided.
What changed recently, how it's actually built (from the code graph), and whether you should care. Free account — no card, no spam.