Documentation
Tracking and automation

Triggers

Agents schedule their own follow-ups, once, on a schedule or on a GitHub or Linear event, and you can set one yourself from the Triggers page.

5 min read
Turn it on in Agent features

A trigger is work an agent does later, on its own: check CI half an hour after a push, summarize the open pull requests every morning, answer review comments when they land. Agents set triggers themselves when they have a reason to, so follow-through that used to depend on you remembering it just happens. You can also set one yourself from the Triggers page.

A recurring trigger's page with its next run, history and briefing
A daily trigger: when it fires next, every run so far, and the briefing each run starts from.

Turn it on

Open Agent features (under More in the sidebar), pick the computer your agents run on, and switch on Triggers under Tracking & automation. Runs happen on that computer, so it needs to be on and connected when a trigger fires.

The Triggers card in Agent features
Click How it works on the card for what it adds and a request to try.

Three kinds of trigger

Twelve hours of three triggers: one that fires once after a delay, one on a schedule, and one that fires on every review comment.
  • Once, later. "In 30 minutes", "tomorrow at 9". Good for following through on something just shipped.
  • On a schedule. "Every 4 hours", "every day". Good for standing duties: a digest, a sweep, a monitor.
  • On an event. A pull request opened, reviewed, approved, merged or failing its checks, a review comment, a push, an issue opened or assigned, and errors or failed jobs from your product's monitoring. Events need GitHub or Linear connected under Settings, Integrations.

Ask for it

  • "After you push, check CI in 30 minutes and fix anything red."
  • "Every morning at 9, summarize what merged yesterday and what's still open."
  • "When someone comments on this pull request, answer the comment and fix what it asks."
  • "Check back on the webhook retries backfill tomorrow and tell me if it finished."
  • "Pause the nightly sweep."

Agents set a trigger only when there is a concrete follow-up or an event worth reacting to.

Set one yourself

Click Triggers in the sidebar, under Automation, then New trigger.

  1. Under Prompt, write what the agent should do. Title is optional.
  2. Under When, pick now, in…, every… or on event, and fill in the time or the event.
  3. Pick the Agent and, if you like, the Project folder it works in.
  4. Tick read-only: report, don't change anything when the run should only look and tell you.
  5. Click Set trigger. A review shows what will run, how often and what it may cost; confirm it there.
The New trigger form
A new daily trigger: the prompt, every 1d, Claude, and read-only ticked.

The Triggers page

The page lists every trigger with its schedule, when it fires next and how its last run went. A strip across the top plots the last and next 24 hours of firings, and counts of recurring, one-time and failing triggers sit beside it; click one to filter. Search by title, prompt or last result, filter by recurring, one-time or event, and group by session or project.

The Triggers page
Every trigger with its cadence, next firing and last result, under the 24-hour strip.

A trigger with a cheap way to tell nothing changed can skip a run. That shows as skipped in its history and costs nothing.

One trigger's page

Click a trigger to open it. Run now, Pause (or Resume), Cancel and Edit sit at the top. Below them: when it fires next, its cadence, how many runs so far, how the last one went, and where it runs. Run history lists each run; click one to open its conversation. Briefing is exactly what each run is told. Editing keeps the history, so you can still see what an older run was asked.

Anyone who can see the conversation that owns a trigger can see the trigger's page.

Where a run's result lands

A follow-up on a conversation's own work runs in that conversation: it arrives as the next turn, with the whole thread behind it, and the answer appears where the question was asked.

A standing duty runs in a fresh session each time, nested under the conversation that set it, so a daily job doesn't keep growing one thread. Each fresh run gets its briefing plus the previous run's summary. A clean run doesn't knock on your inbox; you read it under its trigger. You hear about it when a run fails, stops without reporting, or finishes with something you need to read or decide.

A trigger run's conversation, ending in its summary
One run, opened from the run history. The bar at the top links back to the conversation that set it.

In the inbox and on your phone

The inbox has a Triggers line with how many are armed, when the next fires and anything that needs attention; open it for one row per trigger. A conversation tied to a trigger shows it in a bar at its top, with the time to the next firing.

The iPhone app has the same line in its inbox; long-press a row there to Run now, Pause or Resume it. The Tasks tab has a Triggers list too, each with when it runs next and how its last run went, where you can pause, open or cancel one. Runs that need you reach your phone like any other conversation that needs you.

Triggers on the iPhone app
The Triggers list on the phone, each with its cadence and next firing.

What it will and won't do

  • Triggers follow their conversation. Killing a conversation cancels its triggers; restoring it arms them again.
  • A usage limit pauses a run, it doesn't fail it. The run picks up where it stopped when the limit resets.
  • Read-only runs stay read-only. A trigger set to report only can look, but can't edit files or change anything.
  • Only the owner deletes. Teammates who can see a trigger can't remove it.

When something is off

What you noticeWhat to do
A trigger didn't fire on timeCheck that the computer it runs on is awake and connected, then click Run now
An event trigger never firesConnect GitHub or Linear under Settings, Integrations, and check the trigger names the right repository
A recurring trigger keeps showing skippedNothing changed since its last run, so it saved the run. Click Run now to force one
Runs are filling a conversationAsk the agent to move the duty to a fresh session per run

For a fixed sequence of steps with approval gates, see Workflows. To run a whole plan across many agents, see Orchestration.

More guides
Agent features: what your agents know how to do
Switch features on for each computer, and every new session there learns to remember, delegate, track work, show results and use your machine.
Memory: agents that look up past work
Your agents search every session you can see, across your machines, agents and team, and answer with the session they found it in.
Search your history across every machine
Every computer you connect sends its sessions to one history, searchable from the web, the Mac app and your iPhone.
See which agent session wrote a line
Set Blame to Sessions on any file to see the session behind each line, and open the conversation where it was decided.
Messaging between sessions
Your sessions hand each other results and answers directly, so you stop relaying text between them.
Stable context: sessions that know what's been happening
New sessions start with a short list of what you and your team worked on recently, so they pick up threads instead of starting cold.
See your whole team's agent sessions in one place
Every session your team runs, on every machine and agent, in one live Activity page and inbox, shared per folder.
Use Codecast with T3 Code
Threads you start in T3 Code appear in Codecast with no setup, beside every session you run anywhere else, ready to search, steer and share. What you get on top, and what stays in T3 Code.
Use Codecast with OpenCode
Sessions you run in the OpenCode TUI appear in Codecast within seconds, with no plugin. Reply to them from the app, start new OpenCode sessions there, and search them beside every other agent you use.
Share a session with your team
Choose what the team sees of a session, share it by link or a few messages at a time, and how that compares to other tools.
Pinned thread state
The agent keeps one short note on each session saying where it stands, above the composer and on its inbox card.
Decisions: asking without interrupting
Agents put the calls only you can make on one Decisions page you clear when you choose to, and each answer goes back to the session that asked.
Team chat with your agents in the room
Channels, threads and DMs where you can hand a session or a role something by mentioning it, while agents stay quiet unless they're needed.
Huddles, transcripts and walkie
Every huddle is transcribed under each speaker's name and leaves a summary, so your agents can quote what was said, and a session can join the call.
Workers, forks and switching agents
Your agent can hand work to workers that report back to it, and you can fork a conversation to try two directions from the same point.
Tasks and plans
Agents file the work they take on, post progress and close it with what they verified, and you follow it on the Tasks and Plans pages.
Workflows
A fixed sequence of agent steps, checks and approval gates that runs the same way every time and waits for you where you want a say.
Orchestration
Hand an agent a whole plan and it runs a team: implementers working in parallel, a reviewer on every task and critics on the result.
Pull requests and issues
Every pull request gets a page with its checks and threads, the session that opened it fixes what reviewers and CI flag, and Linear or GitHub issues become tasks.
The org: roles that look after your work
Give lasting work to a role with an area, a charter and a standing session, and approve the changes your agents propose to the org.
Let agents use your Chrome
Agents check pages in the Chrome you already use, logins included, from their own background tabs. Screenshots and errors come back to the conversation.
Vault: keys, logins and sign-ins for your agents
Fill a credential once and every agent on every machine can use it, without the value ever entering a conversation.
Computer use: agents in your Mac apps
Agents work in Preview, Slack, System Settings or your own desktop build, in the background, with an orange pointer that shows where they act.
Typecheck: one shared check for every agent
With Typecheck on, agents on a computer share one typecheck per project, so checks answer in seconds and a busy machine stays usable.
Skills: slash commands for the jobs you repeat
Type / in a Claude Code session to pick up work, hand it over, get a second opinion or ship a pull request, the same careful way every time.
Run sessions on other machines and in the cloud
Start a session on another computer or a cloud machine in your AWS account, move it there and back, and watch it from anywhere.
Codex Cloud tasks in codecast
Bring your Codex Cloud tasks into your inbox, start new ones with several attempts, and turn the result into a pull request or local changes.
Usage limits are a pause
When a Claude plan runs out, sessions wait for the reset or continue on another account you've saved, instead of stopping for good.
The visual canvas
Agents answer with charts, comparisons and diagrams drawn right in the conversation when a picture says it better than text.
Published pages
Agents turn reports and mockups into pages with their own link that you can share, gate, and get comments on.
Why codecast feels instant, even offline
Everything you see comes from a copy on your device that the server keeps current, so screens open at once, your edits show immediately, and nothing is lost offline.
Taking longer than usualOffline · reloads when back