- Tests you approved are unchangedtest/scanner.limit.test.ts✓ Passed
- Tests passpnpm test scanner · exit 0✓ Passed
- Claims match the record4 of 4 confirmed✓ Passed
- Codex tried to break itno failing test found✓ Passed
Vakr in 42 seconds.
Music: “Happy Beats / Business Moves vol. 9” by Sascha Ende (ende.app), CC BY 4.0.
A day with Vakr.
Pick an hour. The window on the right is the app at that moment, and it works: merge the review, answer the agent, halt the runs.
An interactive demo with sample Projects. Nothing you do here is sent anywhere.
And the real thing.
Screenshots of the Mac app, with sample Projects.
Every run shows its work.
An agent saying “done” isn’t evidence. Every review in Vakr carries what was actually run. Point at a check to see what stands behind it.
Tests first
The agent writes failing tests for the task. You read a few short tests instead of a long diff, then approve. Only then does it write the code, and it can’t touch the tests you approved.
What it said, checked against what it ran
“All tests pass” is set against the actual command and its exit code. A contradiction holds the merge.
Another agent tries to break it
An agent from a different vendor gets the diff with one job: write a test that fails. If it can’t, there is nothing for you to read. You ask for it on a review; on Pro it runs on every one.
A track record per Repo
Merged clean, reworked or reverted, taken from your own reviews and shown when you pick an agent.
Budgets per Project
Each Project gets a share of your weekly limit, so one busy night can’t spend the week.
Free on your own Machines.
Everything you need to run agents and trust what comes back is free. Safety is never a paid extra. During early access, all of it is unlocked.
- Sessions, Needs you and dispatch to any agent
- Worktrees, tests first and reviews with checks
- What it said, checked against what it ran
- Budgets, policies and Halt all runs
- Other Machines and your phone
- No caps on agents or runs
- A second agent tries to break every run
- A track record per Repo, and the agent it suggests
- A queue that waits for your limit, and the week plan
- Repeating tasks
- History you can search and ask
- Signed proof of a run you can share
- Everything in Pro, for life
- For the first 200 people from the waitlist
- Nothing to pay now; I’ll write when it opens
No monthly plan, nothing hosted, and Vakr never sells tokens: your agents stay on the plans you already pay for.
Questions.
Ask the way you’d search in the app. Press / from anywhere on the page.
Nothing matches that. Try “phone”, “main” or “agents”.
What is Vakr?
A Mac app with a small background service that tracks your Projects and the coding agents working on them. It queues work, collects everything waiting on you into Needs you, and checks each run before you merge.
Which agents does it work with?
Claude Code and Codex natively, and about 22 more over ACP, OpenCode among them. Vakr uses the subscriptions you already have; it never sells tokens and never installs an agent for you.
Will it touch my main branch?
No. Every run gets its own branch and worktree. Vakr never pushes to main, and it never writes to your agent config without asking and making a backup first.
Where does my code go?
Nowhere. Sessions, code and the database stay on your Mac. Machines you pair talk to each other directly over your own network.
Can I use it from my phone?
Yes. Turn on Your phone in Settings › Machines and open the page over Tailscale to answer questions and approve runs.
What will it cost?
Nothing during early access. After that, everything you need on your own Machines stays free, with no caps. Pro is $99 a year and adds a second agent on every run, the track record, the limit-aware queue, repeating tasks and searchable history. The first 200 people from the waitlist can buy a founding licence for $149, once, instead.
How is it different from running Claude Code in a few terminals?
Terminals give you agents. Vakr gives you one list for everything they need from you, a view across every Project, and proof with each review: the tests you approved are untouched, what the agent said matches what it ran, and a second agent can be sent to try to break it.
What does it need?
macOS 15 or later on Apple silicon or Intel, and at least one coding agent installed and signed in. A Linux VPS can join as a Machine.