t3-thread-actions
Skill110 starsChanged 17 days agoMITView on GitHub
Settle, unsettle, archive, unarchive, rename, stop, snooze, fork, switch model or search T3 threads, manage a thread's queue, or resume one after a usage limit. Use for "settle <thread>", "archive <thread>", "rename <thread>", "stop <thread>", "snooze <thread>", "resume after the limit". For "what is <thread> asking", use t3-waiting.
In this repository22 other files it gives its agents.
- AGENTS.mdAGENTS.md
- t3-checkupSkill
- t3-cleanupSkill
- t3-digestSkill
- t3-get-outputSkill
- t3-getting-startedSkill
- t3-handoffSkill
- t3-install-toolsSkill
- t3-new-threadSkill
- t3-phone-answersSkill
- t3-pm-threadSkill
- t3-project-reportSkill
- t3-project-trackerSkill
- t3-save-learningsSkill
- t3-statusSkill
- t3-syncSkill
- t3-templatesSkill
- t3-tots-rulesSkill
- t3-updateSkill
- t3-usageSkill
- t3-waitingSkill
- t3-watchSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 0 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 9
Follow the t3-tots-rules skill (skills/t3-tots-rules/SKILL.md).
- Very specific: line 28
Remove: --cancel <runId|all>, only with the user's OK or for your own stale messages.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Follow the t3-tots-rules skill (skills/t3-tots-rules/SKILL.md). Find the thread with its "Finding a thread" section, then do only the action asked, on that thread.
Actions
| Action | Command |
| Settle or unsettle (reversible) | t3ctl settle --thread '<t>', t3ctl unsettle --thread '<t>' |
| Archive or unarchive (hides it from the sidebar and stops its PR watches; unarchive brings it back, but PR watches stay off until set again) | t3ctl archive --thread '<t>', t3ctl unarchive --thread '<t>'. A running thread can't be archived (THREAD_RUNNING); stop it first only if the user says so. Check t3ctl pending --thread '<t>' first: archiving a thread with an open question or approval stops phone answers for it, so tell the user before archiving one |
| Rename | t3ctl rename --thread '<t>' --title '<new>' |
| Stop the running turn | t3ctl interrupt --thread '<t>'; --hold also holds the queue |
| Stop a run and its sub-tasks | t3ctl stop --thread '<t>'; this holds the queue |
| List sub-tasks, stop one Claude subagent | t3ctl subtasks --thread '<t>', t3ctl subagent-stop --thread '<t>' --subagent <id> |
| Snooze | t3ctl snooze --thread '<t>' --until '+2h' (or HH:MM or ISO), t3ctl unsnooze --thread '<t>' |
| Switch model (idle only) | t3ctl model --thread '<t>' --model <m> [--effort e] [--provider p] |
| Fork to try an idea | t3ctl fork --thread '<t>' --title '<x>' |
Queue
A message sent with --queue waits until the current run ends. Queued messages can get stranded when a run is cancelled or the queue is held.
- List:
t3ctl queue --thread '<t>' - Deliver one into the running run now:
--steer <runId>. Use--steer allonly when every queued message is yours. - Remove:
--cancel <runId|all>, only with the user's OK or for your own stale messages. - Edit or reorder:
--edit <runId> --text '<new>',--move <runId> --firstor--before <runId|end>. - Continue after a hold:
--resume. A held queue is intentional, so ask before resuming one you didn't hold.
Usage limits
t3ctl limitslists threads stopped on a usage limit, with the reset time and whether auto-resume is set.t3ctl resume-at-reset --thread '<t>'makes T3 continue at the reset, or right away if it has passed.t3ctl snooze-until-reset --thread '<t>'hides it until then.--cancelundoes either.
Reply
One line per thread with the new state, read back from t3ctl threads.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.