Skip to content

Integrations · the tool belt

Hook Claude Code up to GitHub, Linear and Slack

…and the rest of your agent crew while you're at it.

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.

Hang GitHub, GitLab, Linear, Slack, Langfuse and MCP on a repository, and choose on each tool's tag what an agent may do with it.

  • macOS & Windows
  • Apache-2.0
  • No account
  • Keys stay encrypted

Workshop · repo atlas

Load the belt

Tap a tool to hang it on the agent's belt. Tap it again to put it back on the board.

  1. GitHub
  2. Linear
  3. Slack
  4. Langfuse
  5. MCP
  6. Sandboxes
  7. GitLab

Connections belong to the repository, so every new workspace in atlas starts with this belt.

7 of 7 tools on the belt

The tags on the belt

Each tool's tag says what the agent can do with it and what stays held for you. Change a setting and the run below changes too.

  • Wrench · a gh account or a token

    GitHub

    Auto-fix CI

    Commit, push and open draft PRs. Merge, or auto-merge when checks pass. Watch CI and reviews. Start from an issue or PR link. Open PRs in GitHub or Graphite.

    A failing CI run waits on Home with Fix CI in… for you to press.

    Auto-fix sends the failing log to the tab you picked: 2 fixes a commit, 5 a PR. Never for review comments.

    Agents in repositories that use it get GH_TOKEN. Turn approvals on to see each push first.

  • Level · a Linear API key

    Linear

    What agents may do

    Agents get no Linear tools. Foreman still reads issues for New workspace.

    Agents can search and read issues, projects and comments.

    Agents can also create and update issues and comment.

    Nothing goes to Linear from an agent.

    They can't create, edit or comment.

    Issue text is written by others, so treat this like any tool that changes things.

    Foreman doesn't change issue status itself.

  • Megaphone · a bot token

    Slack

    What agents may do
    Rule: PR ready → message the reviewer

    Foreman sends your notification rules. Agents can't post.

    Agents may also post to #launches, the channel you allowed.

    Messages to anyone but you wait for your Send.

    This rule has your consent; change any part of it and it asks again.

    Outbound only. Never pings @here or @channel.

  • Magnifier · one project's key pair

    Langfuse

    What agents may do

    Agents get no Langfuse tools.

    Agents get search_observations, get_trace and get_observation.

    They can't change anything in Langfuse.

    Langfuse Cloud EU, US or HIPAA, or your own host over HTTPS.

  • Power strip · Model Context Protocol

    MCP

    Let agents start workspaces

    Every workspace agent gets foreman-app: its status, its checks, a preview, its own chat history.

    No agent can start a workspace or queue a message.

    Agents can hand tasks to other workspaces in atlas and read the replies. Other repositories are refused.

    MCP servers a repository proposes start only after you approve them.

  • Toolbox · experimental, off by default

    Sandboxes

    Where the workspace runs

    E2B · Hosted · 1-hour sessions on Hobby · 2, 4 or 8 vCPUs.

    Daytona · Hosted · 24-hour sessions · 1, 2 or 4 vCPUs.

    Vercel Sandbox · Hosted · 45-minute sessions on Hobby · 2, 4 or 8 vCPUs.

    Docker · This computer, or another over SSH · No session limit · 1 to 8 vCPUs.

    Provider keys never enter the sandbox. Needs a GitHub remote. Claude Code, Codex and OpenCode, not Cursor.

    Turns keep going after you quit, up to the provider's session limit.

  • Pliers · an access token

    GitLab

    gitlab.com or self-managed

    Merge requests and issue links work like GitHub's, on your own GitLab host.

    Fewer events: employee triggers don't see failed pipelines, approvals or requested changes.

    foreman gitlab add https://gitlab.example.com

Example run: ENG-123, start to finish

8 of 8 steps work with this belt. Missing tools grey out their steps.

  1. Linear

    Starts from ENG-123, “Login loops after sign-in”. The agent reads the issue and its comments.foreman run issue ENG-123

    Hang Linear on the belt to start from an issue. Without it, you type the task.

  2. Workspace

    Workspace osaka: its own git worktree on eng-123-login-loop.

  3. Langfuse

    Reads the trace of the failing POST /login: the redirect fires before the session is saved.

    Hang Langfuse to let the agent read your app's traces.

  4. MCP

    Checks itself with foreman-app. Delegation is off, so it writes the regression test too.

    Hang MCP to see what an agent can do with Foreman's own tools.

  5. Pull request

    Pushes the branch and opens draft PR #42 on GitHub.foreman pr create atlas --draft

    Hang GitHub or GitLab to push and open the pull request.

  6. CI

    CI fails: Tests, 1 failing. Foreman saw it, so you don't have to refresh.

    Hang a code host to watch CI.

  7. Back to green

    Auto-fix is off: Home shows CI failed and you press Fix CI in…. Checks pass.

    Hang GitHub to watch CI and fix it.

  8. Slack

    Drafts “PR #42 is ready for review” to @maya, the reviewer. It waits for your Send.

    Hang Slack to tell the reviewer.

Merging stays with you: press Merge, or foreman pr merge atlas --auto to merge when the checks pass. Linear's own GitHub integration moves ENG-123 along. Illustrative run; the names are made up.

Wrench · GitHub & GitLab

What can Claude Code do with GitHub in Foreman?

Everything between the last commit and the merge: commit, push, open a draft pull request, merge or auto-merge when checks pass, and watch CI and reviews, from the window or the foreman command.

  1. Committed
  2. PR
  3. Checks
  4. Merged
  • Draft PRs in one click, opened in GitHub or Graphite. Stacked PRs if you use gh stack.foreman pr create atlas --draft
  • Merge or auto-merge the way the Merge button does, after it shows you the checks and review state.foreman pr merge api --auto --yes
  • Fix CI from Home's CI-failed row, or opt in to auto-fix: 2 fixes a commit, 5 a PR, 20 minutes each.foreman pr auto-fix api on --tab 1
  • Workspaces from a link. An issue fills in the task; a PR link checks out its head too.foreman run issue https://github.com/owner/repo/issues/12

GitLab too: gitlab.com or a self-managed host, with merge requests and issue links. It reports fewer events, so employee triggers miss failed pipelines, approvals and requested changes.

The Editor layout: a file diff in the middle, the changes on the left and the chat on the right
Editor A file diff in the middle, the changes on the left, the chat on the right. Read it before it ships.

Level · megaphone · magnifier

How do Linear, Slack and Langfuse plug into your agents?

Each one is a named connection with its key in Foreman's encrypted storage. Linear and Langfuse give agents tools at the access you choose; Slack sends the notifications your rules describe, and asks before it messages anyone but you.

Linear: start from the issue

foreman run issue ENG-123 opens a workspace with the issue as its task, and foreman linear issues lists what's assigned to you.

Nothing
No Linear tools. Issues still come in for New workspace.
Read only
Search and read issues, projects and comments.
Read and write
Also create and update issues and comment.

Foreman doesn't change issue status itself; Linear's GitHub integration does that.

Slack: eight rules, one Send button

Rules tell you, the reviewer or a channel when…

  • a reviewer is requested
  • a PR is ready for review
  • CI fails
  • changes are requested
  • a PR is approved
  • a PR is merged
  • an agent needs me
  • an agent finishes

Ask before messaging on Slack · Acme

Message @maya about #42 Fix login loop?

SkipEdit messageSend to @maya

Outbound only: you can't approve from Slack. It never pings @here or @channel.

Langfuse: traces, read only

One project's key pair, on Langfuse Cloud or your own host. Agents get a foreman-langfuse MCP server with search_observations, get_trace (errors, cost and latency) and get_observation.

They can't change anything in Langfuse.

Power strip · MCP

What is the foreman-app MCP server?

It's the socket every workspace agent gets for free: a local MCP server that lets an agent check its own status, run its checks and, if you allow it, hand tasks to other agents. No install, no open HTTP port.

  1. Socket 1 · Always on

    foreman-app, in every workspace

    get_workspace_status, get_check_results, run_checks, open_preview and read_chat_history, held to the agent's own workspace and chat.

    Opt in per repository and agents can also start_task, send_task and wait_for_turn, in that repository only. Branches are never merged for you.

  2. Socket 2 · Needs your approval

    Servers a repository proposes

    A repo can propose MCP servers in .foreman/settings.toml, naming variables, never secret values. No agent gets one until you approve it on this computer; a change asks again.

  3. Socket 3 · Opt-in, new

    Your own Claude Code or Codex session

    Turn on Agents on this computer and a session you run yourself can sign in over MCP on 127.0.0.1, with a six-digit code you type and a level you pick, from Read only to Never ask.

    Honest note: neither setup has been run by hand against a current release yet, and a level bounds what comes through Foreman, not the agent itself.

The tool crib

How do connections work across repositories?

A connection is a named account with a default per service. Each repository inherits the default, picks another, or turns the service off, and a broken connection blocks that service instead of quietly using another account.

  • Keys live in Foreman's encrypted storage. Linear, Slack and Langfuse keys are never passed to agents.
  • Secrets are never command-line arguments: they come from standard input or a prompt that doesn't echo.
  • Sandbox provider keys stay on your computer and never go into a sandbox.
Stock the crib from a terminal
foreman connections add github --gh-account octocat --label Work
printf %s "$LINEAR_API_KEY" | foreman connections add linear --token-stdin --label Acme
foreman connections apply work --repos api,web

Settings → Integrations Illustrative

  • Work · gh account octocatUsed by 2 repositoriesDefault
  • Personal · personal access tokenUsed by 1 repository
  • Acme Linear · key ••••x7QaAgents: Read onlyDefault
  • Foreman bot in Acme3 rules · you are MayaDefault
  • Kite traces · Needs attentionLangfuse refused this key. The repository using it has no access until you fix it; it won't use another account.
Repository settings → Integrations
RepoGitHubLinearSlack
atlasInherit · WorkInheritInherit
kitePersonalInheritOff
docsInherit · WorkOffOff

The rest of the workshop

What else does Foreman do besides integrations?

It runs your agents in parallel, one git worktree each, and adds automations, auto-fix, flows, AI employees and remote control from your phone, all local-first and free.

  • One worktree per agent

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

    The core

  • Automations and schedules

    Save a prompt and run it hourly, daily, weekly or monthly on Claude Code. Schedules run while Foreman runs, with or without a window.

    7 built-in templates

  • Auto-fix CI and auto-review

    A failing CI log goes back to your agent within the budget you set, and a read-only reviewer reads each finished diff.

    Opt-in, with a budget

  • Flows that loop until green

    A graph of agent steps that loops until the checks pass, with gates where you decide. Off until you turn it on.

    11 built-in flows · experimental

  • AI employees

    A role in plain words, a hard hat and git-backed memory. Their Slack posts and Linear changes wait for you.

    15 job templates

  • 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.

    No Foreman relay: same Wi-Fi connects directly, anywhere else goes over your own Tailscale network.

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

The Automations library of templates and commands, with schedules beside it
Automations Templates and commands as cards, with every schedule beside them.

Toolbox · experimental

Can the crew work in an E2B or Docker sandbox?

Yes, as an experiment: a workspace's worktree and agent can run in a sandbox on your own account. It's off by default, in Settings and per repository.

  • E2B1-hour sessions on Hobby
  • Daytona24-hour sessions
  • Vercel Sandbox45-minute sessions on Hobby
  • DockerNo session limit

Needs a GitHub remote. Claude Code, Codex and OpenCode; not Cursor. Provider keys never enter the sandbox, and turns keep going after you quit, up to the provider's session limit.

Other workshops

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. Next to T3 Code, Foreman adds flows, AI employees, scheduled automations, remote sandboxes and a CLI where every command takes --json.

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 day in the workshop

Get started in a minute

Download the app, add a git repository, and give each task its own workspace. Hang the tools when you need them.

  1. Download Foreman

    Free for macOS (Apple Silicon) and Windows. You need Git on your PATH.

    Download Foreman
  2. Add a repo, sign in an agent

    Sign in Claude Code, Codex, Cursor or OpenCode in Settings → Agents. Connect GitHub, Linear or Slack in Settings → Integrations.

  3. Create workspaces

    One per task, each a fresh worktree on its own branch. Review the diff, then open the PR.

Prefer a terminal? One line.
foreman run fix-a-bug "the login page loops after sign-in" --repo atlas --wait

# Start a workspace from a template and wait for the turn. CLI reference

Shop talk

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.

The pegboard has empty hooks.

Foreman is three weeks old: 31 releases since 1.0.0 on 27 October, built in the open under Apache-2.0. Star it to hear about the next tool, fork it, or bring one of your own: a good first issue is a fine place to start.