Automations

An automation is a defined agent job that runs manually, once at a chosen time, or on a recurring schedule. Use one when the instructions and trigger are stable enough that each occurrence can be reviewed as the same kind of work.

Create an automation without choosing a named project, or select an existing project when it should use that project's instructions, memory, files, skills, and connector choices. Workspace connectors are available by default; turn off any the automation should not use. Changing the selected project affects future runs, while earlier results stay in their history.

When an automation uses Computer, files and browser sign-ins follow its project, as they do for chats and dashboards in that project. Use a separate project when work should not share those desktop files. Moving an existing automation to another project can require help to move its desktop; earlier results remain in history.

An automation defines

• the job instructions;
• an optional description that explains the operational purpose;
• the project context and capabilities used by future runs;
• model and reasoning settings;
• a manual-only, one-time, daily, weekly, or other supported recurring schedule;
• enabled or paused state;
• optional execution guardrails such as maximum steps and runs per day;
• run history, last result, and next due time.

The Model menu in New automation and Edit automation lists Light, Medium and High first. Below them, Search models and the model list let you pick Workspace default or a specific model; a specific model also shows its reasoning effort. While your organization's shared Codex subscription is on, or on the Free plan, there is no model list. In Edit automation, Workspace default then sits under the levels, and under Codex a model saved earlier stays until you pick a level or Workspace default. A catalog model choice saves its family and uses the newest eligible version for each new run. An already-running occurrence and its history keep the version used.

Each run appears in the automation history with its result. Assigning an agent to receive results does not grant it project access or share that agent's private conversation or credentials.

Open and manage an automation

Select a card on Automations to open its instructions, schedule, team and run history. Choose View run on the card to open its latest available run.

Cards show the assigned agents, or No agent assigned when none are attached. Use Manage team inside the automation to change assignments and follow-up settings.

Use the switch on the card to enable or disable the automation directly. The actions menu contains Delete automation, which asks you to confirm before removing it.

People and result handovers

Select Manage team to see current members and add or remove people or agents. Use Contact settings to choose someone for questions and failures. An automation can run without an assigned recipient. Assignment records who is responsible and who receives future results; it does not grant project access.

Each run keeps the recipients selected when it starts. When it finishes, assigned agents can receive a result without a new message from you. Keep informed asks for a brief acknowledgement unless you have given standing follow-up instructions. Act on findings asks the agent to assess actionable results under its existing permissions and approval rules. Delivery may wait while an agent is paused, busy, or unable to run under its current access or budget.

The automation does its independent job; the receiving agent carries out your authorized follow-up using its own context and tools, such as its email tools. For example, ask the agent to email you a summary whenever a particular automation finishes. The agent can save your standing instruction in its own project instructions, scoped to that agent and automation, and choose Act on findings for future handovers. The automation does not need to be rewritten to use the agent's mailbox. An existing standing instruction can also authorize follow-up while the subscription remains Keep informed. Receiving a result or a wakeup alone never grants permission to send messages, publish, or take other external actions; existing approval requirements still apply.

An agent can follow an accessible automation, choose Keep informed or Act on findings for itself, and stop following. These changes affect only its own notifications. People can manage the same settings in Manage team. Stopping notifications suppresses pending automatic pickups and attention wakeups; it does not erase already received results or stop work already running. Following again does not replay canceled notifications.

Older reminders saved as ordinary agent wakeups need to be canceled separately in the agent's scheduled wakeups.

An agent can inspect an automation result when it has access to that run. Its Automations panel shows routines and recent findings. A confirmed handover appears in the agent's conversation with an Open run link. A result is shared with an agent only when people who can read that agent's conversation can also access the source run. Otherwise, the result remains in its original run.

Adding someone applies to future runs; old runs retain their original recipients. Removing an agent also stops its pending automatic notifications. Older runs may show No handover recorded.

Agents and other workspace runs can discover and inspect an accessible automation. With edit permission, they can update its name, description, instructions, enabled state, or schedule. An agent that only has permission to view it can inspect its results and manage its own subscription. Editing changes the definition for future runs; it does not start a run or rewrite historical sessions.

Receipt and run outcome are separate

An automation can finish while an assigned agent is still waiting to receive its result. Received confirms pickup, not that the agent understood or acted on the finding. Each recipient has its own status. Delivery does not rerun the automation or change its outcome.

The detail page shows one Latest run with its execution status and a short result excerpt. The recipient's avatar and linked name identify who receives the result. Waiting for Clay to pick it up means Clay is assigned but has not received the handover yet. Received includes the confirmation time. Each recipient has its own state, and an older dispatch without confirmation remains Unconfirmed. A run with no agent shows No agent assigned and remains available to people with access.

Open a run to see its recipients beneath the result. Select Nudge Clay to request pickup of that run. Nudge sent means the request was accepted; Waiting for Clay to pick it up remains until receipt is confirmed. Repeated clicks do not create repeated pickups. The nudge uses the agent's model budget and does not rerun the automation or change its schedule. Paused agents must be resumed first, and current access and budget rules still apply.

If the pickup is canceled or can no longer proceed, Nudge stopped replaces the waiting message. The result remains available for the agent's next eligible turn; selecting the nudge again does not start another turn. Received always means the agent has confirmed pickup, even if that turn later stops.

The agent can open the result for more detail. Viewing recipients does not wake an agent, and a nudge does not authorize external messages or publication.

Latest run and history

Latest run follows the newest execution, including work still running. History lists earlier executions without repeating their reports. Each entry shows its run outcome and agent handovers. Select Open run or Open to inspect the full result in that run's conversation. Loading more history does not change which run is featured.

A running execution shows Result pending until its own result is available. Older runs may show No handover recorded. If handover details cannot be loaded, use Retry; the run can still be opened.

Create and test an automation

You can ask an agent to propose an automation while chatting with it in d5s. Review the instructions and schedule in its suggestion, then select Create. The confirmation is shared with teammates: reopening the conversation or confirming the same suggestion again opens the same automation. The originating agent follows it with Keep informed and receives a short creation reference. The automation still runs with its own instructions and project context.

An older suggestion may ask you to request a new one to avoid duplicates. You can still dismiss the old card for everyone.

Write instructions that make one occurrence self-contained: identify the inputs to inspect, the decision or artifact to produce, the destination, and what should happen when there is nothing to report. Review the inherited workspace connectors and turn off those the project does not need.

Start with Manual and use Run now while refining the definition. After the output is reliable, add a one-time or recurring schedule. Run now behaves like a scheduled run.

Each run that starts uses the configured model.

Prefer an agent when responsibility is ongoing

Automations optimize for a repeatable job with independent run records. Agents optimize for a persistent role and conversation. If the work should notice context, decide what responsibility is due, and continue a relationship across check-ins, staff an agent instead. If the only useful result is a maintained view, consider a dashboard.

Google workspace protection

Google content can be used for the work you authorize in this workspace. Tasks using AI may send relevant content to the AI providers used by the workspace. Disconnecting does not erase information already retrieved. See Google data protection for AI processing and stored-data guidance.