Skip to content
RepoStaff

Coming soon · AI staff for the team you already have

Rent the AI staff. Own the brain.

Hand it a backlog ticket. It will hand back a pull request with proof attached. You merge.

See pricing

Your backlog in.
Pull requests out.

How it will work

  1. You

    Describe the outcome,
    or hand it a ticket

  2. RepoStaff

    Writes and
    tests the code

  3. You

    Review
    and merge

Example ticket · illustration

Follow “Add CSV export” through.

You step in 3 times. The default. A person approves the plan, approves the pull request, and merges it.

  1. You

    Start the ticket

    Add the ready label to the issue.

  2. Marshall

    Check it's ready

    Not a duplicate, right repo, enough detail to start.

  3. Marshall

    Write the plan

    Plan and acceptance criteria go on the ticket.

    You approve the planIn the Management UI or a chat button
  4. Worker

    Build it

    Fresh VM. Branch, commits, draft PR.

  5. Worker

    Prove it

    Runs your checks, attaches proof to the PR.

    CI passes
    A separate Marshall session signs off
  6. Marshall, then you

    Review the PR

    Marshall reviews before you do.

    Marshall reviews
    You approve on GitHub
  7. You

    Merge

    Ticket closes. Learnings are recorded.

    You mergeOn GitHub or in the Management UI

Every preset: a fresh VM per ticket, CI before merge, and CI config, infra, migrations and .env files off limits.

Boss rule

Merge isn't deploy.

RepoStaff stops at merge. Agents get no production credentials.

The proof on a PR supports your review. It doesn't guarantee the code is right.

House rules

  • Policy lives in policy/workflow.yaml.
  • Every change is a commit in git log.
  • Agents read a pinned snapshot.
  • Chat proposes. It doesn't authorize.

01The backlog has plans. Give it capacity.

Mission brief

Built by a staff engineer, for software teams.

The situation

Worthwhile features, finite capacity. What the AI staff learns lands in a repo you own.

Built for

5–100 engineers on GitHub, already using AI coding tools, whose seniors are stretched thin.

02The AI staff you rent.

Roster

Pillar 1

Ephemeral agents

Prove it. Then disappear.

  1. boot
  2. build
  3. prove
  4. gone

One task, one fresh VM. No production credentials.

Pillar 2

Long-term agents

Marshall stays. One per org.

Tasks come and go. Marshall stays across all of them and keeps what it learns.

Plans, triages and checks every task.

Pillar 3

Management UI

The receipt.

#1291 CSV export
merged
Gates passed
3 / 3
Cost
per PR

Illustration. Stop or redirect any run.

mainThe brain feeds all three. It isn't rented. It's a git repo in your org.

03Buried in AI-slop issues and PRs?

Incoming

Marshall will take the first pass, so your seniors only open what's worth their time.

Yes, an AI staff company is offering to clean up after AI. I see the irony.

[in]5 new · illustration

  • #1288App crashes sometimes
  • #1289Login broken on Safari
  • #1290Improve performance
  • #1291Add CSV export
  • #1292Switch auth library
  • Ready1

    • #1291 · Criteria draftedAdd CSV export
  • Needs more2

    • #1288 · Asked for repro stepsApp crashes sometimes
    • #1290 · Asked which issueImprove performance
  • Duplicate1

    • #1289 · Same as #1203Login broken on Safari
  • Needs a person1

    • #1292 · Auth goes to a personSwitch auth library

01 · Triage

Signal from noise.

Duplicate, wrong repo, too thin to start, or ready.

02 · Sort

Your labels, not ours.

It applies the labels you already use. Sorting never starts work by itself.

03 · Supplement

Fill in what's missing.

Repro steps, related links, acceptance criteria.

It labels and comments. Your people close. Every pass runs on your model key.

04The brain is a repo.

Your save file

Inventory

$ git clone git@github.com:your-org/brain.git
  • org/how your company works
  • decisions/why things are the way they are
  • memories/what Marshall learned, and where
  • skills/what your agents know how to do
  • policy/who has to approve what

Rules of the save

Every memory records its source. When a person and Marshall disagree, the person wins. Cancel and you keep the brain.

05Five commits to ready.

Quest log

  1. 01

    Select repos and a brain.

    Install the GitHub App. Create or connect a brain repo.

  2. 02

    Add your key.

    Anthropic or OpenAI, on your own account. No resale.

  3. 03

    Enroll compute, tracker, chat.

    One runner, one tracker, one chat.

  4. 04

    Marshall reads. You confirm.

    Confirmed answers land in the brain, with their source.

  5. 05

    Label it ready.

    Your ready label starts the run.

06Hosted management. Your compute.

Map

Coming soon: RepoStaff will host the Management UI. Your agents will run on compute you choose.

  1. Your Linux/KVM hardware

    Coming soon. Old machines in a closet count. Each task gets a fresh VM.

  2. Your AWS account

    Coming soon. One CloudFormation stack you review and create. Cost estimate on the pricing page.

  3. RepoStaff-hosted AWS

    Coming soon. After self-run. Waitlisted, no date.

Check your stack.

Tap what your team uses. Choose one per group to check fit.

Code and CI

Tickets

Chat

Infrastructure

Cloud compute is planned for customer AWS accounts after Linux/KVM hardware. Azure and Google Cloud are not supported. All integrations are pre-launch.

Your stack is coming soon.

GitHub, Linear, Slack, AWS. These integrations are planned, not operational.

Get started

Your Anthropic or OpenAI API key. No subscription logins. No token resale or markup.

07Flat per org. No seats.

Shop

Planned prices; they may change. Model usage on your own key and your own compute are billed separately. Full pricing

Coming soon

Self-run

$199

per org, per month

Your compute. No limit on members or repos. 14-day trial.

Get started
Coming soon · Waitlist, no date

RepoStaff-hosted compute

$499

per org, per month

RepoStaff's AWS. 100 VM-hours included, then metered up to your spend limit.

Hosted compute waitlist details
Coming soon · If support allows

Self-hoster rate

$49

per org, per month

Up to three members on their own hardware.

Self-hoster rate details

Your backlog is waiting.

Good fit: 5–100 engineers · Code on GitHub · Already using AI coding tools · US-based

Not supported yet: Jira · GitLab or Azure DevOps · Regulated enterprise · Data residency guarantees

Get startedHosted compute waitlist details

Code location, model keys, SOC 2, Devin and more: read the FAQ.