agentleFS
DeepusleepyDeepusleepy/t3-tots/skills/t3-new-thread/SKILL.md

t3-new-thread

Skill110 starsChanged 17 days agoMITView on GitHub

Start a new T3 thread or send a follow-up into one. Use for "new thread", "/new", "start a thread in <project>", "follow up in <thread>", "send this to T3".

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 0 are distinctive, and 4 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.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Follow the t3-tots-rules skill (skills/t3-tots-rules/SKILL.md), including "Sending into threads".

    line 7
  2. Very specific:

    If the prompt is longer than a few lines, write it to a file and pass --prompt-file <file>.

    line 17
  3. Very specific:

    Add --queue only when the user wants the follow-up to wait for the run to end.

    line 22
  4. Very specific:

    Leave out --wait for long coding tasks, say it's running, and offer t3-watch.

    line 23

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Follow the t3-tots-rules skill (skills/t3-tots-rules/SKILL.md), including "Sending into threads". Send only the prompt the user asked for.

1. Resolve (read-only)

  • Computer: the one the user names, else the one their message came from. For a follow-up, the computer that owns the thread.
  • Project: t3ctl projects. --project takes an id, title or path. Titles and paths ignore case; ids must match exactly. If the folder isn't a project yet, ask before ~/.t3/bin/t3 project add <path> (Windows: t3.cmd project add <path>).
  • Thread, for follow-ups: t3-tots-rules "Finding a thread".
  • Model: t3ctl models lists the models already used. Aliases are haiku, sonnet and opus, and pick the newest of that family; other ids look like claude-sonnet-5-5, gpt-6.1-sol or grok-4.7. Effort is low, medium, high, xhigh or max. With none given, the project default applies.

2. Write the prompt

  • Pass on the user's words and context. Leave the design decisions to the thread.
  • If the prompt is longer than a few lines, write it to a file and pass --prompt-file <file>.

3. Send

  • New thread: t3ctl new --project <P> --model <m> --effort <e> --prompt-file <file> [--title '<t>'] [--mode plan] [--attach a.png,b.jpg] [--wait --timeout 300]
  • Follow-up: t3ctl send --thread '<id|title>' --prompt-file <file> [--attach ...] [--queue] [--wait [--timeout 300]]
  • Add --queue only when the user wants the follow-up to wait for the run to end.
  • Leave out --wait for long coding tasks, say it's running, and offer t3-watch.
  • t3ctl new records the thread in started-here.json, so t3listen arms Resume at reset if the thread later stops on a usage limit.

Fallback on macOS

Use the UI only if t3ctl fails and the user is OK with T3 taking focus: t3ui new '<project search>' '<model search>' '<Low|Medium|High|Extra High|Max>' '<prompt>' or t3ui send '<exact sidebar title>' '<prompt>'. Then check the result with t3ctl read. If macOS says the app isn't allowed assistive access, tell the user which app to allow in System Settings > Privacy & Security > Accessibility.

Reply

In 1 or 2 lines: the thread title and id, project, model and effort, and the reply or "running". Say whether the message was steered into a running run, queued, or not delivered.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.