How to run Claude Code on a schedule in 2026: 8 ways compared

Disclosure: Enjoy is our product, and it runs Claude Code. We checked every way below against its maker’s own pages on October 28, 2026, and we say where another way is the better pick, including the ones built into Claude Code.

The ways to run Claude Code on a schedule, compared

A scheduled run is Claude Code working without you: a prompt you save once, started at the times you choose, such as a code review every weekday at 9:00 or a dependency check every Monday. Anthropic builds three ways into Claude Code, and its docs say when to use each: “Use cloud tasks for work that should run reliably without your machine. Use Desktop tasks when you need access to local files and tools. Use /loop for quick polling during a session.” The other ways start Claude Code from your CI, a server you run, or another app.

Way	Best for	Price, checked October 28, 2026	Where it runs	Your computer on?	Agents
Cloud scheduled tasks	Work that runs with your laptop shut	Included in Claude Pro ($20 a month), Max (from $100), Team and Enterprise	Anthropic’s cloud, on a fresh clone of a GitHub repository	No	Claude Code
Desktop scheduled tasks	Recurring work on your own files	Included in the same plans	Your computer, in the Claude app	Yes, with the app open	Claude Code
/loop	Checking on a deploy or a pull request while you work	Included in the same plans	Your computer, in an open session	Yes, with the session open	Claude Code
GitHub Actions	Scheduled jobs in a repository’s CI	Your Claude plan or an API key, plus GitHub’s runner minutes, free in public repositories	GitHub’s runners	No	Claude Code
cron, with claude -p	A server that’s always on	Free; runs on your Claude plan or an API key	Any computer you run	Yes	Claude Code, or any command-line agent
Paseo	Many agents on schedules, free	Free	The machine running its daemon	Yes, that machine	Claude Code, Codex, Copilot, OpenCode, Antigravity, Muse Code and more
Orca	Many agents on schedules, free	Free	Your computer, or an Orca server you run	Yes, that machine	Claude Code, Codex and any other coding agent
Enjoy	Any lab’s agent, across all your projects	A monthly subscription, free for 7 days, with your phone and other computers (pricing)	Your computers	Yes, with Enjoy open	Claude Code, Codex, Grok Build, Antigravity, Muse Code, OpenCode

Prices are per month as each maker lists them in the US, before tax. Claude Code itself needs a paid plan or an API key: Claude’s pricing says it “is included in all paid plans,” from Pro at $20 a month, or $17 a month billed yearly, and it “shares the same usage limits as the rest of your plan.” Whatever starts it, a scheduled run draws on those limits unless it runs on an API key.

Cloud scheduled tasks, formerly routines

Best for: work that should run with your laptop shut, on a GitHub repository.

A cloud scheduled task, formerly called a routine, “is a saved Claude Code configuration: a prompt, one or more repositories, and a set of connectors, packaged once and run automatically.” It runs on Anthropic’s servers, “so they keep working when your laptop is closed.” Create one at claude.ai/code/routines, in the Claude desktop app, or by typing /schedule in Claude Code, as in /schedule daily PR review at 9am. Besides a schedule, a task can start from an API call or a GitHub event, such as a new pull request. Each run opens as a session you can read afterward, and it starts in auto mode, where “a classifier model allows or blocks actions that would otherwise wait for your approval.”

Price: “Scheduled tasks are available on Pro, Max, Team, and Enterprise plans,” and they “draw down subscription usage the same way interactive sessions do.” Pro is $20 a month, and Max starts at $100.

Watch for: “Scheduled tasks are in research preview. Behavior, limits, and the API surface may change.” “Each repository is cloned at the start of a run,” so a task works on what’s in the repository, not on the files on your computer: Anthropic’s comparison lists its access to local files as “No (fresh clone).” Custom schedules can’t run more than hourly, as “Expressions that run more often than once per hour are rejected,” and a bare cron expression is in UTC. When your plan’s limit runs out, “additional runs are rejected until your usage window resets,” unless you’ve turned on usage credits.

Desktop scheduled tasks, in the Claude app

Best for: recurring work on the files and tools on your own computer, when Claude is your only agent.

In the Code tab of the Claude desktop app, select Scheduled, then New task, and choose a local folder and a schedule: hourly, daily, weekdays, weekly, or a cron expression. Or ask in any session, as Anthropic’s example does: “set up a daily code review that runs every morning at 9am.” Each run is a new session that can “edit files, run commands, create commits, and open pull requests,” with the permission mode you gave the task, and in a Git project it can work in its own worktree, away from your unfinished changes. If your computer slept through a run, the app starts “exactly one catch-up run for the most recently missed time” when it wakes.

Price: included with Claude Code in every paid plan, from Pro at $20 a month. The app runs on Mac and Windows, with Linux in beta.

Watch for: “Tasks only run while the desktop app is running and your computer is awake.” Its Keep computer awake setting stops idle sleep, but “Closing the laptop lid still puts it to sleep.” A task in Manual mode that needs a tool it hasn’t been allowed “stalls until you approve it,” so Anthropic suggests clicking Run now after creating a task and choosing “always allow” for each prompt. It runs only Claude.

/loop, in an open session

Best for: checking on a deploy, a build or a pull request while you work.

Type /loop 5m check the deploy in Claude Code, and the prompt runs every five minutes. Leave out the interval, and Claude “chooses one dynamically,” waiting anywhere from a minute to an hour between runs. For a single reminder, say it in plain words: “remind me at 3pm to push the release branch.” Behind both are Claude Code’s cron tools, which take a five-field cron expression in your local time.

Price: included in Claude Code.

Watch for: “Tasks only fire while Claude Code is running and idle. Closing the terminal or letting the session exit stops them firing.” “Recurring tasks automatically expire 7 days after creation,” there’s “No catch-up for missed fires,” and a daily task starts 0 to 30 minutes late on purpose, so that every session’s tasks don’t call Anthropic at the same moment.

GitHub Actions

Best for: scheduled jobs on a repository, run by your CI, with no computer of yours involved.

Anthropic’s GitHub Action runs Claude Code on GitHub’s runners. Its docs say “With a prompt input, the Claude Code GitHub Action runs in automation mode on any GitHub event, including a cron schedule,” and their example runs every day at 0 9 * * *. It signs in with a Claude API key, or with a token from claude setup-token that uses your Claude plan.

Price: “If you authenticate with an OAuth token, runs use your Claude subscription instead of API billing.” The runners use your GitHub Actions minutes: GitHub’s billing page says standard runners are free in public repositories, and GitHub Free includes $12 of compute a month for private ones, at $0.006 a minute on a standard Linux runner.

Watch for: “GitHub runs scheduled workflows only from the default branch and, in public repositories, disables the schedule after 60 days without repository activity.” With a plain prompt, “Claude has no shell or GitHub API access until you grant the tools the prompt needs.” And an OAuth token “is tied to the subscription of the person who ran claude setup-token.”

cron, with claude -p

Best for: a server or a computer that’s always on, with nothing between your scheduler and Claude Code.

The do-it-yourself way: have cron, launchd or Windows Task Scheduler start Claude Code without its interface. “To run Claude Code in non-interactive mode, pass -p with your prompt and the CLI options you need,” say Anthropic’s docs, as in claude -p "Find and fix the bug in auth.py" --allowedTools "Read,Edit,Bash". On a machine where you can’t sign in through a browser, claude setup-token makes a one-year token to set as CLAUDE_CODE_OAUTH_TOKEN, and the docs say to “Pass --permission-prompts none when nobody is available to answer permission prompts, for example in a scheduled job.”

Price: free. The token “authenticates with your Claude subscription and requires a Pro, Max, Team, or Enterprise plan,” or you can use an API key, billed by the token.

Watch for: everything around the run is yours to build, from logs to telling you when it’s done. “Without --bare, a -p session runs the hooks in a project’s .claude/settings.json and connects the servers in its .mcp.json, even in a folder you’ve never trusted,” and bare mode “does not read CLAUDE_CODE_OAUTH_TOKEN,” so it needs an API key. In Claude Code’s issue #97074, open since September 25, users report that headless claude -p uses “~1.8× more of the 5-hour window per token” than the same work in the terminal, with no reply from Anthropic there; our memo on Claude’s limits has the details. And Anthropic’s terms say Pro and Max limits “assume ordinary, individual usage of Claude Code and the Agent SDK.”

Paseo

Best for: scheduling Claude Code, Codex and other agents for free, on a machine you keep running.

Paseo is free and open source, under Apache 2.0 on GitHub, and runs “Claude Code, Codex, Copilot, OpenCode, Pi, Antigravity, and Muse Code agents” through one interface. Its schedules start “a new agent for you on a cron cadence: at this time, run this prompt, in this repo, with these agent settings,” and its heartbeats send “a recurring prompt back into one existing agent so it can reassess and continue the same conversation.” You set them up in the app, by asking an agent, or with paseo schedule create on a headless box.

Price: free.

Watch for: “Paseo runs a local server called the daemon that manages your coding agents,” and “Agents run on your machine,” so the machine running the daemon has to stay on. Paseo’s homepage says you can run it “on a home lab, or a cloud machine.” Each run’s workspace and agent are archived when it ends, unless you turn off Archive on finish in the app: “The CLI and MCP schedule tools do not expose this option.”

Orca

Best for: scheduling Claude Code, Codex and other agents for free, with a server of your own for when your laptop sleeps.

Orca is “Free and open source for macOS, Windows, and Linux,” under the MIT license, and runs “Claude Code, Codex, and any other coding agent in parallel.” Its automations “run a prompt on a schedule,” created with orca automations create or on its desktop Automations page. A schedule can be a preset such as weekdays, “plus cron expressions or RRULE strings,” and a precheck lets you “Skip scheduled work when a cheap shell probe fails,” such as when there’s no pull request to review.

Price: free.

Watch for: an automation runs on the machine that hosts it, which its docs call “where each automation is stored and scheduled.” For agents that keep going after your laptop sleeps, Orca’s ways to run point to a Remote Orca Server on “an old laptop, Mac mini, home server, cloud VPS, or team box,” and “Orca does not sell managed VPS hosting.”

Enjoy

Best for: scheduling any lab’s agent across all your projects, without writing cron or keeping a terminal open.

Enjoy is a desktop app for Mac, Windows and Linux that runs Claude Code, Codex, Grok Build, Antigravity, Muse Code, and local models through OpenCode, on the plans you already have. A workflow “is a set of instructions you save once and run whenever you need it, or on a schedule.” What it adds for scheduled work:

• Any agent, on a schedule. Set Repeat to every day, every weekday, every week, or a cron expression, in the time zone you choose, and pick the agent, model, effort and permissions its runs use. One workflow can run Claude Code and the next Codex or a local model.
• Each run, a conversation. Every run starts a new conversation that you can read, reply to and steer, at your desk or from your phone.
• Every project at once. Each project keeps its own workflows, and JOY keeps an eye on all of them: ask it what needs you.
• A computer that stays on. Scheduled runs happen on the computer that has the project, so a Mac mini at home can run them all night while you follow along from your laptop or phone.

Price: Enjoy is one monthly subscription, and the first seven days are free, with a card to start. It includes your phone and any browser and your other computers, and each teammate you invite needs a subscription of their own. See pricing.

Watch for: scheduled workflows run only while your computer is awake with Enjoy open: “Enjoy must be running,” and “The computer must also be awake.” “Missed times are skipped, not made up,” and when the previous run “is still working or waiting for your answer, that scheduled time is skipped.” Enjoy runs Claude Code headless, so if the reports in issue #97074 are right, Claude Code in Enjoy uses the five-hour window faster than in a terminal. Enjoy isn’t open source, and it’s a monthly subscription on top of your AI plans once the free week ends.

When Anthropic’s own scheduling is enough

If Claude Code is the only agent you use, start with what’s built in. Cloud scheduled tasks cover the nights your laptop is shut, Desktop scheduled tasks cover your own files, and /loop covers the afternoon you spend waiting on a deploy, all at no cost beyond your plan. Cloud tasks can also start on GitHub events and API calls, and Desktop tasks make up the latest missed run when your computer wakes, which Enjoy doesn’t.

Look further when you want Codex or another lab’s agent on the same schedules, a job in your CI, or a server you control. OpenAI’s agent has a scheduler of its own: in the ChatGPT desktop app, scheduled tasks “can work with local projects,” and OpenAI says to “Keep the computer on and the app running when a scheduled task needs local files.”

How we compared

We read Anthropic’s Claude Code docs, pricing and terms, GitHub’s Actions billing, Paseo’s and Orca’s sites, docs and GitHub repositories, Claude Code’s issue #97074, and OpenAI’s guide to scheduled tasks on October 28, 2026, and every one is linked below. We quote each maker’s own words for its limits and leave out what it hasn’t documented. Scheduling in these apps changes often, so we recheck this guide against its sources at least once a month, and the date at the foot of the page says when we last did.