You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
[Feature request] Run an agent automation/prompt on auto settled thread
#16820
Every time a pull request I worked on in T3 Code merges, the same chore follows. I open Jira, find the ticket, drag it from In Review to Done, and try to remember how many hours the work took. The thread that did the work already knows all of this. It knows which ticket the branch came from, what changed, and when I started. I'm the one copying that knowledge into another tool.
I'd like to hand that step to an agent. Next to "Settle threads when their pull request merges", I'd write one prompt for the project, for example "use /jira-close to move this ticket to Done and log the time", and pick the provider and model to run it. From then on, a merge that settles a thread also starts that prompt in a side chat of the thread. The side chat starts with the thread's context, so it can find the ticket and work out the hours without me explaining anything. The thread itself stays settled and out of my way. If the side chat fails, I get a notification.
Jira is my case, but the prompt is whatever the team needs. Another team might close the Linear issue and post a summary of the change, or have the agent write the changelog entry from what the thread built. T3 Code already runs a shell command on settle. This gives the same moment to an agent, which can do the follow-up work that needs judgment and the thread's context.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
Uh oh!
There was an error while loading. Please reload this page.
Every time a pull request I worked on in T3 Code merges, the same chore follows. I open Jira, find the ticket, drag it from In Review to Done, and try to remember how many hours the work took. The thread that did the work already knows all of this. It knows which ticket the branch came from, what changed, and when I started. I'm the one copying that knowledge into another tool.
I'd like to hand that step to an agent. Next to "Settle threads when their pull request merges", I'd write one prompt for the project, for example "use /jira-close to move this ticket to Done and log the time", and pick the provider and model to run it. From then on, a merge that settles a thread also starts that prompt in a side chat of the thread. The side chat starts with the thread's context, so it can find the ticket and work out the hours without me explaining anything. The thread itself stays settled and out of my way. If the side chat fails, I get a notification.
Jira is my case, but the prompt is whatever the team needs. Another team might close the Linear issue and post a summary of the change, or have the agent write the changelog entry from what the thread built. T3 Code already runs a shell command on settle. This gives the same moment to an agent, which can do the follow-up work that needs judgment and the thread's context.
All reactions