Skip to content

Dispatch office · open while Foreman runs

AI coding agent automation, dispatched to the right hard hat.

Foreman is a free, open-source desktop app for macOS and Windows that runs AI coding agents in parallel. Claude Code, Codex, Cursor and OpenCode each work in their own git worktree on their own branch, while you review diffs, answer questions and open pull requests from one window.

CI fails, a pull request merges, Linear assigns an issue: each event goes to the crew member you chose, in a worktree of its own.

  • macOS & Windows
  • Apache-2.0
  • No account
  • v1.2.4

Dispatch docket · illustrativeNo. 0412

Event
CI failed on #42 fix-login-redirect
Rule
Auto-fix CI, opted in for this PR
Crew
osaka, tab 1 · Claude Code
Budget
fix 1 of 2 this commit · 1 of 5 this PR
Also
Slack → you (ci-failed rule)

Try it · repo atlas

The dispatch board

Send an event, or let them arrive. Each one is routed by the rules on the right to a hard hat, a worktree and a lane. Switch a rule off and watch where the work goes instead.

Send an event

You

Clock

Channel

GitHub

Git

Linear

Memory

Departures

atlas · Europe/London
Events routed this morning, newest first
TimeEventCrewTrackStatus
09:38@PRIYAPriya#rate-limits · planASKS
09:31MERGED #57Theodocs-after-57 · freshARRIVED
09:20LINEAR ENG-311Ravieng-311-session-timeoutARRIVED
09:12CI FAILED #42Auto-fixosaka · fix-login-redirectARRIVED
09:00SCHEDULE 09:00morning-triagekyoto · triage-2026-11-18ARRIVED

Ticket No. 41509:38

@Priya in #acme

  1. @Priya add rate limiting to atlas
  2. Priya reads the repo and asks first; nothing runs until you approve her plan

Priya#rate-limits · plan

Lands in: Needs you

Arrived
an agent or employee took it
Asks
it needs an answer from you
Notified
a message or a card; nothing started
No route
logged, nothing started

Routing rules

Always routed, no rule needed: you asking, schedules and @Priya.

Branch & PR

  1. Draft PRNo. 414

    #58 docs: rate limits from #57

    Theo pushed docs-after-57 and opened a draft pull request.

  2. Draft PRNo. 413

    #63 ENG-311: sessions expire too early

    Ravi pushed and opened a draft pull request.

  3. PushedNo. 412

    #42 · fix pushed, CI running again

    Fix 1 of 5 for this pull request · Foreman never claims CI recovered; the PR watch reports it.

  4. BranchNo. 411

    kyoto · triage-2026-11-18

    +46 −0 · 9 issues labelled · workspace kept

Slack · outbound

  1. SentNo. 412

    To you · CI failed on #42

    fix-login-redirect · 2 failing checks. Messages to yourself send without asking.

Needs you

  1. QuestionNo. 415

    Priya has 2 questions before she plans

    Per user or per API key? And what should a blocked request get back?

    Send answersPlan with what you have

  2. Held SlackNo. 414

    Post to #eng: “Merged #57: rate limits”

    A message to anyone but you waits for Send unless the rule has your consent.

    SkipSend

  3. Held LinearNo. 413

    Move ENG-311 to In Review?

    Linear changes wait for you. Foreman doesn’t change issue status by itself.

    SkipApprove

bots/e-3fq9.tomlThe rule behind the selected route

[[triggers]]
kind = "event"
source = "linear"
on = ["assigned"]      # polled every 5 minutes
team = "ENG"
label = "bug"
repo = "atlas"         # where the work happens
prompt = "Fix the issue with a test and open a draft PR"
debounce = 60
enabled = true

One Slack rule8 events · outbound only

$ foreman slack rules add --event pr-merged \
    --recipient channel --channel eng --repos atlas
# Mode "ask" by default: each message waits for Send or Skip.
$ foreman slack rules consent 3fa91c2b
# Prints exactly what would go out, and to whom, and asks.

Timetable

What can start an AI coding agent in Foreman?

You, a schedule, @Priya, and events: failing CI (with auto-fix), pull request events, the base branch moving, a Linear assignment and a memory commit. Events start AI employees and the opt-in auto-fix; plain templates and schedules have no event triggers.

Every event Foreman hears, whether it can start an agent, and what else reacts
EventStarts an agent?What else reactsHow it's heard
You ask: the app, the CLI or your phoneYes: a new workspace, with any agent you signed inYou choose the agent and the branchWhen you ask
A scheduleYes: a template or a flow, on Claude CodeSkipped, not replayed, if Foreman wasn’t runningOn the clock, in your time zone
@Priya in a channelYes: Priya plans; the parts start after you approveHer questions come first, on one cardWhen you post
CI fails on a pull requestYes, if you opt in: auto-fix sends the logs to your tab. Employee trigger ci-failedSlack rule ci-failed · a row on HomePR watch
A PR is merged, gets a review request, changes requested or approvedYes: AI employee triggers merged, review-requested, changes-requested, approvedSlack rules pr-merged, review-requested, changes-requested, pr-approvedPR poll, every 2 minutes at most
New review commentsNoA notification. Send a comment to a tab with a clickPR watch
The base branch movesYes: employee trigger base-movedBase-moved notice (opt-in): never rebases, merges or starts an agentEvery 5 minutes
A Linear issue is assignedYes: employee trigger linear, by team and labelThe issue is still there for New workspacePolled every 5 minutes, only while a trigger exists
A memory-repo commitYes: employee trigger memory, by pathOnly commits Foreman made; an employee’s own never wakes itAt once
A turn finishesA read-only reviewer, if you opt in to auto reviewChecks after a turn run commands, never an agentAfter 2 idle minutes
  • Unattended means Claude Code. Schedules, employee runs and flow steps that write code run on Claude Code today; an approval they hit is denied and the turn stops.
  • Guards against loops. Events within a debounce (60 s by default) become one run; at most 3 event runs per item in 10 minutes.
  • GitLab hears less. Merges and review requests reach employee triggers; failed checks, approvals and requested changes don't yet.

Arrivals

Where does the work land?

On a branch in its own worktree first, then as a pull request, a Slack message or a card in Needs you. Anything that goes to other people or changes Linear waits for your button.

  1. Platform 1

    A branch and a diff

    Every workspace is its own git worktree on its own branch. Agents never step on each other, and your main checkout stays clean. You read every change as a diff, and checkpoints rewind any turn.

  2. Platform 2

    A pull request

    Since 1.2.4, AI employees push and open draft pull requests without waiting. You merge, or auto-merge when the checks pass, and open PRs in GitHub or Graphite. GitLab merge requests work too.

  3. Platform 3

    Slack, outbound only

    Notification rules for 8 events. A message to anyone but you waits for Send or Skip unless the rule has your consent. Nothing comes back from Slack; you answer in Foreman.

    • review-requested
    • pr-ready
    • ci-failed
    • changes-requested
    • pr-approved
    • pr-merged
    • agent-needs-me
    • agent-finished
  4. Platform 4

    Needs you

    Questions, gates, held Slack posts and Linear changes, each as a card with the exact preview. Nothing that leaves your computer has a keyboard shortcut.

An organization's Team page: its employees, what each needs from you, and a memory change waiting for approval
Team Each employee, what each needs from you, and a memory change waiting for approval.

The rest of the yard

What else does Foreman run?

Dispatch sits on top of the everyday app: agents in parallel worktrees, flows that loop until green, control from your phone, remote sandboxes and a command line for scripts.

Radio · remote control

Run it from anywhere

Pair your phone with a QR code to approve, steer and comment on diffs, live. Away from home it connects over your own Tailscale network, with no Foreman relay. A browser on another computer can open the whole app.

Same Wi-Fi connects directly. The phone reads terminals but can't type into them, and can't answer an employee's held actions.

Native apps

Today: pair your phone with a QR code and add the phone view to your Home Screen.

Track 01

One worktree per agent

Every workspace is its own git worktree on its own branch. Agents never step on each other, and your main checkout stays clean.

Tiles shows several workspaces side by side so you can step in where you are needed. Built for 10 to 12 agents running at once.

Track 02

Flows that loop until green

Write a way of working down once as a graph of agent steps: loops that run until the checks pass or a review comes back clean, and gates where you decide.

11 built-in flows. Experimental: Settings → Experimental → Flows.

Track 03

Remote sandboxes: E2B, Daytona, Vercel, Docker

Run a workspace's worktree and agent in a sandbox on your own E2B, Daytona, Vercel Sandbox or Docker account. Provider keys never enter the sandbox, and turns keep going after you close the app. Experimental.

Claude Code, Codex and OpenCode, not Cursor. Needs a GitHub remote. Experimental.

Track 04

Agents that use Foreman too

Every workspace agent gets a foreman-app MCP server to check its status, run checks and, if you allow it, hand tasks to other agents. A Claude Code or Codex session you run yourself can sign in to Foreman over MCP.

The MCP server isn't provided to agents in remote sandboxes.

Track 05

GitHub, GitLab, Linear, Slack, Langfuse

Open, merge and auto-merge pull requests, watch CI and reviews, start work from a Linear issue, send Slack notifications and read Langfuse traces. Each connection gets the agent access you choose: nothing, read, or read and write.

Slack is outbound only. An AI employee’s Slack posts and Linear changes wait for you.

Track 06

A real command line

Almost everything in the window works from a terminal with the foreman command. Every command takes --json and exit codes are stable, so scripts and CI can rely on them.

Exit codes are the same for every command.

Other depots

How is Foreman different from Conductor and T3 Code?

Foreman is open source under Apache-2.0 and runs on Windows as well as macOS, while Conductor is a closed-source Mac app. Only Foreman lists AI employees, scheduled automations and auto-fix CI.

A dash means not listed on their site

Foreman compared with Conductor and T3 Code
FeatureForemanConductorT3 Code
Open sourceApache-2.0NoMIT
macOSYesYesYes
WindowsYesMac onlyYes
One git worktree per agentYesYesYes
AgentsClaude Code, Codex, Cursor, OpenCode, any ACP agentClaude Code, Codex, CursorClaude Code, Codex and more
Flows that loop until checks pass, with human gatesYes——
AI employees with git-backed memoryYes——
Scheduled automations and auto-fix CIYes——
Your own cloud sandboxes (E2B, Daytona, Vercel, Docker)Experimental——
Scriptable CLI, --json on every commandYes——
Approve from your phoneYes—Yes
No account, no telemetry uploadYes——
From each project's own site as of November 2026. “—” means it isn't listed there, not that it's missing. Spot a mistake? Tell us.

First shift

Get started in a minute

  1. Download Foreman for macOS (Apple Silicon) or Windows. Free, no account. You need Git on your PATH.
  2. Add a git repository and sign in an agent with your own Claude Code, Codex, Cursor or OpenCode login or key.
  3. Create workspaces, one per task. Then add rules: auto-fix on a pull request, a Slack rule, or an employee with a trigger.
Download ForemanFree · v1.2.4 · macOS & Windows

Or dispatch from a terminal

Start a workspace from a template and wait for the turn

foreman run fix-a-bug "the login page loops after sign-in" --repo atlas --wait

See what is waiting for you

foreman approvals

Open a draft pull request

foreman pr create atlas --draft

Every command takes --json. CLI reference

Enquiries

Questions developers ask

Something missing? Open an issue or read the guide.

What is Foreman?

Foreman is a free, open-source desktop app for macOS and Windows that runs AI coding agents in parallel. Claude Code, Codex, Cursor and OpenCode each work in their own git worktree on their own branch, while you review diffs, answer questions and open pull requests from one window.

Is Foreman free and open source?

Yes. Foreman is free and licensed under Apache-2.0, with the full source on GitHub at github.com/elite-build/foreman. There is no paid tier, no account and no Foreman cloud service.

Which coding agents does Foreman support?

Foreman runs Claude Code, Codex, Cursor and OpenCode, plus any agent that speaks the Agent Client Protocol (ACP). Each agent signs in with your own subscription or API key, set in Settings → Agents.

How do I run multiple Claude Code agents in parallel?

Add a git repository to Foreman and create one workspace per task. Each workspace is a separate git worktree on its own branch, so several Claude Code sessions edit the same repository at once without conflicts. Tiles shows them side by side.

Why does each agent need its own git worktree?

A git worktree is a second checkout of the same repository on a different branch. Giving each agent its own worktree means agents never overwrite each other's files, every change lands on a reviewable branch, and your main checkout stays untouched.

Does my code leave my computer?

Only through the agents you choose. Foreman runs locally with no account and no telemetry upload; your agents talk to their own model providers. In your own workspaces, an approval marks any command that leaves your computer before it runs, and AI employees hold Slack posts and Linear changes for you.

Do coding agents need my approval for every command?

Only if you want them to. Approvals are a setting: with Claude Code tool approvals on (Settings → Account), each command is spelled out part by part before it runs, anything that leaves your computer is marked, and Always allow shows exactly what it adds. Every agent still works in its own worktree, and checkpoints rewind any turn.

Which operating systems does Foreman run on?

Foreman runs on macOS (Apple Silicon, signed and notarized) and Windows. You need Git on your PATH and a login or API key for at least one agent.

Can Foreman run coding agents on a schedule?

Yes. Any saved prompt template can run hourly, daily, weekly or monthly in your time zone, and opt-in auto-fix sends failing CI logs to your agent. Scheduled runs use Claude Code and happen while Foreman is running, which can be without a window or from login.

Can I control Foreman from my phone?

Yes. Pair a phone with a QR code to answer approvals, steer agents, comment on diffs and read terminals, live. On the same Wi-Fi it connects directly, and anywhere else through your own Tailscale network. Native iOS and Android apps are coming soon.

Can coding agents run in a cloud sandbox like E2B?

Yes, as an experiment. A workspace's worktree and agent can run in a sandbox on your own E2B, Daytona, Vercel Sandbox or Docker account, for Claude Code, Codex and OpenCode. Provider keys stay on your computer and never go into the sandbox.

Which integrations does Foreman have?

GitHub (pull requests, merges, CI and reviews), GitLab, Linear, Slack and Langfuse, plus MCP. Each connection has a default per service, each repository can pick its own, and you choose what agents may do with it: nothing, read, or read and write.

How is Foreman different from Conductor?

Foreman is open source under Apache-2.0 and runs on Windows as well as macOS, while Conductor is a closed-source Mac app. Foreman also drives OpenCode and any ACP agent, and adds flows that loop until checks pass, AI employees and a scriptable command line.

How is Foreman different from T3 Code?

Both are open-source GUIs for coding agents. Foreman adds flows (agent graphs that loop until tests pass, with human gates), AI employees with a git-backed memory, a foreman CLI where every command takes --json, phone approvals and remote sandboxes.

Can I script Foreman or use it in CI?

Yes. The foreman command starts workspaces, sends messages, answers approvals, opens and merges pull requests and runs flows. Every command takes --json and exit codes are the same for every command.

Now boarding: contributors.

Foreman is three weeks old with 31 releases since 1.0.0 on 27 October, built in the open under Apache-2.0. Star it to hear about the next one, fork it, or take a good first issue.