Skip to main content
NovaCore · the governed agent platform

Nova does the work. You make the calls.

Assign it the job — a bug, an incident, a report, the backlog. Nova works on machines you own, asks when it is unsure, and nothing consequential happens without your approval.

  • runs on your infrastructure
  • GitHub · Jira · Slack
  • nothing ships without you
mission m-2841 · checkout
@nova
  1. 09:00 Ticket #2841 assigned to @nova — duplicate confirmation emails queued
  2. 09:00 Step leased by builder-01 — a machine you own leased
  3. 09:05 Bug reproduced, failing test written working
  4. 09:47 Asked a question on the ticket — then waited, for free waiting
  5. 11:38 Fixed — 214 tests pass, 3 files changed done
  6. 11:39 PR #912 opened for review review
approval requested click one — this one is live

Deploy checkout to production — rollback ready, 30-minute watch after.

  1. 14:02 Approved by you — policy check passed approved
  2. 14:32 Shipped — watched 30 min, nothing moved verified
cost $1.87 elapsed 5h 32m
Why it exists

The chat session ends. The mission doesn't.

Work assigned to Nova is durable state, not a conversation. Every mission survives restarts, model swaps and weekends — and resumes exactly where it stopped, evidence attached.
The problem

Why AI pilots stall in engineering teams.

The work never actually lands.

A chat session ends and so does the work. Nothing was finished, nothing was tracked, and nobody can pick it up tomorrow.

Your security team says no.

And they are right to. Handing production access to something nobody can govern, audit or reverse is not a risk anyone should sign off.

It cannot reach your systems.

Your repositories, test databases and internal services are not on the public internet, and they are not getting a firewall exception.

NovaCore was built to solve those three first, and the AI part second.

What you get

Four things you can hold us to.

01

Work that gets finished

Assign a ticket, get a reviewed pull request with tests. Nova keeps going across days, waits for people without burning budget, and picks up exactly where it left off.

02

Nothing ships without you

Anything that touches production comes to you first — what it wants to do, why, what it affects, and how to undo it. One click to approve, one to reject.

03

Your code never leaves

Agents run on machines you own, inside your network. We never need access to your systems, your VPN, or an inbound firewall rule.

04

Proof, not promises

Watch any mission live. See what it changed, what it cost, and how often your team had to correct it — per team and per repository.

Staying in control

You approve anything that matters.

Every change that reaches production arrives as a plain-English summary. No dashboards to learn, no logs to read.

It tells you what it wants to do, why it thinks that, what it affects and how to undo it — then waits. In the browser, on the pull request, or in Slack. Whichever you already use.

If you approve it, Nova ships and then keeps watching. If anything looks wrong in the next half hour, it puts things back on its own.

↑ The same card, live in the console above. You can reject it there — Nova stands down.

Waiting for you needs approval

Ship the fix for duplicate confirmation emails

What it wants to do
Ship a fix for duplicate order confirmation emails. 3 files changed.
Why
Ticket #2841. It reproduced the bug, wrote a test for it, and all 214 tests pass.
What it affects
The checkout service, in production. No spending involved.
How to undo it
Roll back to the previous release. It will watch error rates for 30 minutes after and undo it automatically if anything looks wrong.
Approve Reject Ask a question See the work
How much rope

You decide how far it can go.

Five levels, set per team, per repository, and per kind of work — tighter for anything near production, looser for a dependency bump. A setting, not an instruction it might ignore.

Agency level set per team · per repo · per kind of work
  1. L1 Watch only Nova reports what it would have done. Changes nothing at all.
  2. L2 Suggest It writes up the fix and the reasoning. A person does the work.
  3. L3 Draft it It does the work and opens a pull request, then stops.
  4. L4 Ask me first It can ship — once you approve. Where most teams stay. most teams
  5. L5 Go ahead It ships inside limits you set, on work you nominated.
One name on the work

@nova, wherever your team already is.

Nova is the name on the work. Mention it in a Slack thread, assign it a ticket in Jira, review its pull request on GitHub — it is the same agent, on the same mission, with the same record.

Not a new place to work, and not a chatbot persona to manage. A name your team already knows how to talk to, in the tools they already use.

@nova in Slack assigned in Jira reviewed on GitHub answered in Teams cc’d in email watched in the cockpit
Where it runs

Your code never leaves your network.

Nova works on machines you own, behind your own VPN. They reach out for work; nothing reaches in. No ports, no vendor VPN account, nothing for your network team to sign off.

Runtime topology outbound-only below the line
Ingress & egress
GitHub Slack Jira webhooks egress

The only place external credentials live.

Control
mission kernel scheduler policy approvals API

Decides, records, schedules. Never executes.

Execution
engagement host your computers workspaces coding agents

Everything needing an environment runs on your hardware.

Substrate
event log read models memory knowledge artifacts

Row-level isolation, per-tenant encryption.

Outbound only

The machine polls for work over ordinary web traffic. Nothing ever connects inward.

The machine is the boundary

Your checkouts, build caches and dev credentials live there and nowhere else.

Gone when you say so

Uninstall the agent and nothing of yours was anywhere else.

Fits your setup

Works with what you already use.

Where your work lives
GitHub GitLab Azure DevOps Jira Linear
Where your team talks
Slack Microsoft Teams Email
Which AI does the work
Claude Code Codex Cursor your own keys

Bring your own AI accounts and pay your own rates. When a better coding agent arrives — and one will — you switch it in settings rather than in a migration.

Getting started

Your first ticket, this week.

  1. 01

    Connect a repository

    Takes minutes. Read-only to begin with.

  2. 02

    Nova reads your codebase

    And proposes how to build, test and deploy it.

  3. 03

    You confirm what is production

    The one thing it is never allowed to decide alone.

  4. 04

    Assign it a ticket

    And watch the first one run.

See Nova run on your own code.

A working session rather than a slide deck. Bring a repository and a ticket you would otherwise assign to someone, and watch it go from the ticket to a pull request you approve.