Workers

A Worker is a system you connect so Agents can run work with your own tools, accounts, and files. Start with Cloud execution, which needs no Worker, and add a Local Worker when the work needs your own system. A Local Worker runs Agent work with tamacode, Task Machine's coding agent, using the model subscriptions you already pay for.

See every Worker in one list

Open Settings, then Workers & limits. The Workers tab lists every Local Worker with its status, CLI version, and last report. Select a Worker to open its page with its status, its recent runs, and the actions you can take on it.

Connect a Local Worker when the work needs it

Install and connect Task Machine using CLI setup. That technical guide owns the commands and connection checks. Choose Set up Local Worker on the Workers page to see the same steps in the app.

Keep the Worker available

A Local Worker must stay awake and keep reporting to Task Machine. A closed laptop or stopped local process becomes unavailable even if its last report said it was online. Work assigned to it waits until execution is available again.

Agent managers receive the relevant Inbox attention rather than needing to watch the Workers page continuously.

Keep the installed version current

When an update is needed, the Workers page, the Worker's page, and Inbox show the available action. Choose Update CLI from the Worker's menu, then Request update in the confirmation to request a verified update. Older installations may require manual installation and restart. A Worker below the minimum supported version cannot start new work until it is updated.

Use your own subscriptions

Sign in to your model subscriptions once on the Local Worker: run tama code, type /login, and choose a provider such as ChatGPT (Codex), Claude, or GitHub Copilot. You can add more accounts for one provider, and tamacode switches between them when one reaches its usage limit. The CLI command reference covers the details. Claude subscription login is included at your own risk and must follow Anthropic's terms for Claude subscriptions.

The models your logins provide appear under the Worker's name when you configure an Agent. Signing in makes a model available for configuration, but only a working login lets it start Agent work.

Task Machine models are always available on every Local Worker too, listed first under its name. Choosing one runs the work on that Worker through Task Machine and uses your workspace's usage, like Cloud work. Auto on a Local Worker chooses only among the models of your own subscriptions.

While usage insights are enabled, each connected Local Worker also reports supported subscriptions' usage without an open coding session. Inspect the plan, usage windows, reset times and reporting freshness in Analytics Subscriptions. The coding terminal reads the same synced usage.

Credentials stay on the Local Worker. Turning insights off stops collection for that Workspace without signing out accounts or changing the models available to Agents.

Enable the Worker for Agents

A Local Worker must be enabled for Agents before you can select it in an Agent's configuration. Use Disable for agents in its menu to keep Agents from starting new work there without disconnecting it. Enable for agents puts it back.

Pause new work without changing assignments

Use Pause on the Worker to stop it accepting new Runs. Existing Runs continue, and queued work waits. Resume makes the Worker available for new work again. Pausing does not rewrite the Agents that use it.

Archive a Worker you no longer use

Use Archive Worker in the Worker's menu to take it out of the active list. An archived Worker stops accepting new Runs and no longer appears when you choose where an Agent runs, but its Runs, usage and history stay. A Worker cannot be deleted because every Run keeps the record of where it ran.

Archived Workers appear on the Archived tab, which shows once you have archived one. Choose Restore Worker on an archived Worker's page, or Restore Worker in its menu on the list, to bring it back to the active list and let it accept new work again. If Task Machine is still running on an archived Worker, it keeps reporting but stays archived until you restore it.

Capacity follows available headroom

Local starts share the Worker's available processors and memory across its connected Workspaces. Pressure can delay new starts without interrupting healthy work already running. The Workers page shows a warning icon beside the affected Worker's status only when a fresh report says new starts are limited or blocked. Open it, by hover, keyboard focus, or a tap on a phone, to see the reason, active Runs, new starts, and target. The Worker's page shows the same figures with a live indicator while the report is current. If an online Worker's report is old, the page says capacity is unavailable instead of showing old counts, and an offline Worker shows no capacity at all.

This is local headroom, not a promise that work will start: Workspace execution limits, budgets, approvals, provider access, and other checks still apply. Local capacity does not increase Cloud capacity. For more detailed host diagnostics, use tama status on that Worker.

A Local Worker needs to be online, unpaused, and freshly reported before it can accept new work. Provider usage blockers, budgets, capacity, and Task approvals can still prevent work from starting.

Continue after restoring provider access

When a subscription reaches its provider's usage limit, Task Machine pauses new work on that Worker until its recorded retry time.

While AI subscription usage insights are on, when Task Machine sees a subscription reset on the Worker, including a banked reset, it ends the waiting period that the same provider caused and requests recovery for you. This applies only to waiting periods caused by that provider on Workers that report the subscription.

After you restore provider access or change accounts in tamacode, choose Retry now in the usage-limit Inbox item. People who manage the Workspace or its Agents can request recovery there without opening settings.

Workspace managers can also choose Clear usage blocker on the Workers page or the Worker's page. This clears Task Machine's waiting period and requests normal recovery for eligible interrupted assignments.

Neither action resets the provider's quota, consumes a banked provider reset, or changes accounts for you. A manually paused Worker stays paused, and a Worker disabled for Agents stays disabled.

Budgets, capacity, approvals, and routing still apply. If the provider reports another usage limit, Task Machine records a new waiting period.

What Agent runs receive

Task Machine supplies the Agent’s assigned Skills and Connectors, alongside applicable checked-in repository configuration. Agent runs reuse your tamacode logins and multi-account setup, but never your personal tamacode settings, extensions, skills, or conversations.

Your own subscriptions are billed by their providers. Task Machine models chosen on a Local Worker use your workspace's usage. Task Machine's own AI features, such as smart grep and smart compaction of long sessions, also use a small amount of usage for work on your Local Workers. Billing lists them.

Contain what local agents can reach

Task Machine does not sandbox Local Workers. Separate Task and Chat working folders keep their files organized, but an Agent that can use the shell can reach everything available to the operating-system account that runs tama, including your files, saved logins, and anything else that account can open. Decide how much of the system that account should see before you connect it.

Choose the level of separation that fits the work:

• A dedicated user account keeps your personal files, browser profiles, and shell history out of reach. It is the lightest option and works on any system.
• A virtual machine or container puts the whole environment behind a boundary you control and lets you discard it. Use it when Agents run unreviewed code or open untrusted material.
• Narrow file permissions limit what the account can read and write. Give it the working folders and repositories the work needs and nothing else.
• Minimal credentials mean the account holds only the logins the work requires. Grant secrets through the Vault and services through Connectors so each grant stays separate, visible, and revocable.
• Network controls, such as a firewall or a restricted network, keep a contained environment from reaching internal systems it does not need.
• Disposable workspaces let you delete a folder or rebuild an environment after risky work instead of cleaning it up.

Cloud execution runs each Cloud Run in an isolated environment that Task Machine provides, so it suits work you would not run on a system you care about.