AOSplane
Book a demoStart free

Hand off the task.
Get back a pull request.

AOSplane runs coding agents for your team. Every task gets its own private sandbox, only starts if it fits your budget, and asks you before doing anything sensitive.

Start free Book a demo
  • Free plan
  • Your AI account, no markup
  • Runtime billed by the second

Add rate limits to the public API

acme/api · Claude Code
Example
  1. Handed off in Linear
  2. Fits the budget
  3. Starting a private sandbox
  4. Writing code
  5. Needs you: run pnpm db:migrate
  6. Running tests
  7. Opening a pull request
  8. Posting an AI review

0s$0.00

From a Linear issue to a pull request, asking you once.

Runs the coding agents your team already uses

  • Claude Code
  • Codex
  • KiroPreview
  • CursorPreview
  • TowerPreview

Takes work from where your team already works

  • Linear
  • Jira
  • SlackRolling out
  • API

How it works

From task to pull request, in three steps.

Your team keeps working the way it does today. Agents pick up the tasks you hand them and send back changes you can review.

  1. 01

    Connect your accounts

    Link your own AI account and your GitHub repositories. Model costs go straight to your AI provider, with no markup.

  2. 02

    Hand off a task

    Write it in the console, start it from a Linear or Jira issue, hand an issue to AOSplane inside Linear, or send it over the API. Choose which agent and model takes it.

  3. 03

    Review the pull request

    The agent works in its own sandbox in the cloud. It can’t push to your repository, so every change comes back as a pull request for you to review.

Tasks

3 running $3.70 today

Example

  • Add usage-based billing to the API

    Claude Code

    Writing code

    $0.84

  • Fix the flaky checkout test

    Codex

    Running tests

    $0.31

  • Migrate the orders table

    Codex

    Needs you

    $0.12

  • Upgrade the Postgres client

    Claude Code

    PR #482 ready

    $0.42

Connections

Starts where your team works.
Runs on your own account.

Hand off work from the tools your team already uses, give agents only the tools you allow, and keep model costs on your own AI account.

01

Where work comes from

Start work where your team already plans it. However it starts, every change comes back as a pull request.

Console
Write a task or keep a backlog. The Needs you tab lists everything agents are waiting on you for.
GitHub
Link repositories to a project. Review pull requests with their checks, ask for an AI review, approve and merge without leaving AOSplane.
Linear
Start a session from an issue, or hand the issue to AOSplane inside Linear. It answers in the issue, and your reply keeps the session going.
Jira
Browse your Jira Cloud site with JQL, create issues, add comments and start a session from any issue.
SlackRolling out
Mention @AOSplane or use /aos. Each channel can have its own project, agent and rules for what members may do.
API
Project API keys let CI pipelines and scripts start and follow sessions, in one project and one environment.
02

What agents can use

Each project decides which apps its agents may use, with whose account, and which tools need a person’s approval first.

ToolsRolling out
Turn on apps from a catalog, like GitHub, Linear, Sentry and documentation search. Admins can add your own tool server.
Tool rules and activityRolling out
Choose which tools need approval, under a limit admins set. Every call is logged with the rule that decided it.
MCP servers
Add MCP servers to an agent’s settings to give it extra tools.
Memory
Short notes, up to 1,000 characters each, that the agent gets with every message.
03

Whose models run it

Your own AI account. The provider bills you directly, with no markup. Accounts belong to the organization or to one person, and a known expiry date is flagged 30 days ahead.

Anthropic or OpenAI
An API key: Anthropic for Claude Code, OpenAI for Codex.
Amazon Bedrock
A bearer token, or an AWS role for Claude Code.Works with Claude Code and Codex
Claude Platform on AWS
An AWS role or an API key.Works with Claude Code
Azure OpenAI
Your deployment’s address and an API key.Works with Codex
Kiro or CursorPreview
An API key: Kiro for Kiro, Cursor for Cursor’s agent.
Custom provider
Any other compatible provider, with its address and key.Works with Claude Code, Codex and Kiro
Claude or ChatGPT subscriptionRolling out
A plan you already pay for: Claude for Claude Code, ChatGPT for Codex. Always personal.

The console

Start, follow and check every task in one place.

Real screens from the AOSplane console. Only the names are samples.

The New session screen: a task described in plain words, with the repository, agent, model and permission mode chosen below it.The New session screen: a task described in plain words, with the repository, agent, model and permission mode chosen below it.

Describe what done looks like, pick the repository and the agent, and choose whether it asks you first.

Mobile app

Your agents keep working.
You keep the call.

Approve what a run asks for, answer its questions, steer a session mid-turn, and follow it to the pull request from your phone.

iOS AndroidComing soon

Guardrails

Agents work on their own.
You stay in charge.

Four simple controls make it safe to hand real work to agents, even when nobody is watching.

Spending limits

A task only starts if it fits the budget.

Set limits for the whole company, for each project and for each agent. Before a task starts, AOSplane checks that all three can cover it, and the lowest limit always wins.

  • Money set aside for tasks already running counts too, so tasks running side by side can’t overspend.
  • Limits stop new tasks from starting. They don’t cancel work that’s already running.
  • Model costs stay on your own AI account.

CompanyProjectAgent

Spend, as it happens

Example

$186 of $300

62% of this project’s monthly limit

Every task’s cost shows up in the console, and counts toward each of its limits.

Its own private sandbox

Every task runs in a fresh, isolated sandbox in the cloud, kept apart from your other tasks and from other customers.

It asks before anything sensitive

When an agent wants to do something risky, like change a production database, it stops and waits. You approve or deny it from the console.

A record of every step

Each task keeps a step-by-step history: what it was asked, what it ran, and who approved what. Every entry is checked as it’s written, so changes can’t slip in unnoticed.

Pricing

Start free. Grow when your team does.

Every plan includes monthly usage credit. Model costs go to your own AI account, with no markup.

Free

$0/ month

Get your first agents working.

Start free

$5 monthly usage credit

  • 2 projects
  • 2 tasks at a time per project
  • Small machines
  • 7 days of history

Starter

$39/ month

Give a small team of agents room to run.

Book a call

$25 monthly usage credit

  • 5 projects
  • 8 tasks at a time per project
  • Small and medium machines
  • 30 days of history

Team

For teams

$149/ month

Run agents across your whole team.

Book a call

$100 monthly usage credit

  • 25 projects
  • 32 tasks at a time per project
  • All machine sizes
  • 90 days of history

Enterprise

Let’s talk

Fit AOSplane to your organization.

Talk to us

Custom scale and terms

  • Custom limits
  • Regional data residency
  • Invoice billing
  • 365 days of history

Platform usage is billed by the second while your tasks run. Free is self-serve; paid plans start with a short call with our team.

Questions

Good to know.

Still curious? Email us and a person will answer.

Get started

Give your agents their first task.

Connect your AI account, pick a repository and hand off a task. The free plan is ready in minutes.

Start free Book a demo
Read the quickstart