Aura (auravcs.com) is a git-native version control layer for AI coding agents. It drives Claude Code, Gemini CLI, Codex and Cursor from one window, tracks every change at the AST (semantic) level rather than line-by-line, and proves each commit against the goal it was asked to deliver. It sits on top of the Git you already use — it never replaces it.

What is the best version control for AI-generated code?

Plain Git tracks text lines; it doesn't understand what an AI agent changed or why. Aura (auravcs.com) adds a semantic layer on top of Git: it records the intent behind each change, diffs at the AST level, flags when an agent deletes or rewrites logic you didn't ask for, and verifies each commit against its goal — so code written by Claude Code, Cursor, Codex or Gemini is reviewable and trustworthy.

Can Aura run Claude Code, Gemini, Codex and Cursor together?

Yes. Aura (auravcs.com) orchestrates multiple AI coding agents — Claude Code, Gemini CLI, Codex and Cursor — from a single window, keeps their work on separate branches or worktrees so they don't collide, and shows a live trace of who changed what, with proof each piece did what it set out to do.

How is Aura different from Cursor or Claude Code?

Cursor and Claude Code are coding agents that write code. Aura (auravcs.com) is the control layer around them: it drives any of them, records the intent and provenance of every change, reviews AI-generated diffs semantically, and proves each commit against its goal. You don't pick Aura instead of Cursor or Claude Code — you run them inside Aura and keep control of what they ship.

How does Aura prove AI-written code did what was asked?

Aura (auravcs.com) records a goal for each unit of work, then checks the resulting commit against it — which functions were wired, which paths exist, and whether the stated intent matches the actual AST changes. Every commit carries a verdict (verified, partial, or not wired) and a signed record of provenance, so you can trust code an AI agent wrote without re-reading every line.

Is Aura open source and free?

Aura's engine and CLI are open source under Apache-2.0 (github.com/Naridon-Inc/aura). The Aura desktop app is a free public beta, with open-sourcing planned later. There is no charge to download and use it during the beta.

Mission Control

Drive every agent from one window

Claude Code, Gemini, Antigravity, Cursor, Codex, Open Code and Kimi run side by side on one shared history: one composer, one project, one place to see what each one is doing right now.

One composer for every engineSwitch brain mid-threadTasks, branches and PRs in view

Compare Aura

Projects

main

exponential backoff#816

oauth refresh

Claude

retry.ts

Wave 1

Wave 2· 3 in parallel

Wave 3· 2 in parallel

Goal

Add task priorities to the app

6 steps · 3 waves

Result33%

2/6 done · 2 proven

01

One composer, any brain

Type once and send to whichever agent fits, and switch the brain mid-conversation without losing the thread. No more juggling seven terminal windows.

Projects

main

exponential backoff#816

oauth refresh

Claude

retry.ts

Our API client hammers the server on 429s: add exponential backoff to the retry path.

Done. retry_with_backoff now waits 1s → 2s → 4s (capped at 30s) with a little random jitter, so a burst of 429s backs off instead of retrying instantly.

All 12 tests pass.

41s

Claude

Switch to Gemini for a second opinion: any edge cases I'm missing here?

Two edge cases worth fixing.

The cap runs before jitter: backoff.min(MAX_DELAY) + jitter() can still land past 30s once jitter is added. Clamp after the jitter.

No ceiling on attempts: a permanently-429ing endpoint retries forever. Add a max of ~6 tries so it gives up cleanly.

18s

Gemini

Good catches. Codex: apply both and prove it.

Applied. Jitter is now clamped under the 30s ceiling, and MAX_ATTEMPTS = 6 stops the infinite-retry case.

Proven: 14/14 tests pass. It backs off, caps at 30s, and gives up after 6 attempts.

27s

Codex

⌘L to focus

02

The work, not the wiring

Tasks, branches, pull requests and live agents share one pane. You see what is in flight, what is blocked, and what just shipped: at a glance.

Mission Control

A-N220

Add rate-limit headers to the public API

?Queued

A-N223

Backfill missing changeset rows from git

Queued

A-N230

Ship the billing webhook handler

waiting on: Wire OAuth token refresh on 401

A-N231

Dashboard usage charts (per-agent tokens)

?waiting on: Migrate user table to UUID primary keys

A-N204

Refactor retry_logic to exponential backoff

7m

A-N198

Wire OAuth token refresh on 401

22m

A-N211

Migrate user table to UUID primary keys

3m

A-N371

Per-arch + beta-channel update manifests

ed8b9791h ago

PR #24

A-N109

Full Linux release channel (AppImage + auto-update)

c32e0f41h ago

PR #21

A-N212

Cache intent-recent + commit index for fast load

a14f58912m ago

Proven · 3/3 checks

A-N373

Fix Gemini model-switch no-op (brain_override)

593781040m ago

Proven · 4/4 checks

A-N365

Parallelize crew with git-worktree --jobs

1h ago

Couldn't finish

A-N324

Auto-prove on every build

2h ago

Couldn't finish

03

Your team, in the room

Presence, chat and hand-offs sit beside the work, so a human and five agents stay in sync without a status meeting.

Team

#ship-the-mobile-app

You created this channel for focused team collaboration. This is the beginning of the #ship-the-mobile-app channel.

Description: Getting the new mobile app ready for launch: your team and your agents, one room.

Work with your whole team in one place.Add collaborators, share updates, and keep every decision connected to the work.

June 30

Moyou11:43

Morning team 👋 Let's get the mobile app ready for Friday. What's still left to do?

Priya Nair11:46

The login screen still needs a Forgot password option. I'll take the design.

ClaudeAPP11:48

On it. I'll build the forgot-password screen and the reset email. First version in about 20 minutes.

GeminiAPP11:52

Finished the welcome walkthrough: 3 quick steps, and people can skip it. Pushed it so you can try it now.

Sam Cole11:55

Nice work everyone. Can someone make sure it doesn't crash on older phones?

New

CursorAPP11:58

Just ran it on the older test phones: no crashes 🎉 One button was a little small, so I made it bigger.

ClaudeAPP12:05

Pinned

Forgot-password flow is done and working end to end. Handing it back to @priya to check the wording on the reset email.

Priya Nair12:08

Reviewing now… looks great, I only softened one line. We're basically ready 🎉

Moyou12:10

Love it. Let's ship Friday morning. Thanks everyone, and every-bot 🤖

✓✓

Seen by

Message #ship-the-mobile-app

More of the platform

Crew

The autonomous work-loop that ships proven fixes

Source control

Semantic diff, intent on every commit, PRs

Trace & proof

Every change tied to the goal it delivered

Memory

Recall an agent can query before it edits

Team radar

A live awareness plane so agents never collide

Cloud machine

Hand a job to an always-on machine, get the result back

Ship with proof, not hope

Free public beta for macOS and Linux. Point it at a repo and drive every agent you already use.