agentleFS
smkg75smkg75/wikidoc/skills/quick-tri/SKILL.md

quick-tri

Skill2 starsChanged 18 days agoMITView on GitHub

Quick filing pass over the inboxes alone (Desktop, Downloads) - a full tri pass sized on them, so nothing else in the corpus is read first. Use when the user asks to sort the Desktop or Downloads, or to clear the inboxes.

The file

Files
SKILL.md

Quick tri

The quick pass: skills/tri/SKILL.md run with the argument inboxes, so collect selects the files of the inboxes and nothing else, and the vision barrier at route waits on them alone. Read that skill and follow it from ① to ⑥; the invariants, the delegations and the questions are the same, and this one only fixes the size and the order.

  1. Size. Pass inboxes to collect. One inbox named in $ARGUMENTS: size the pass on that inbox's file count alone.
  2. A pass already hanging. A bench/routing.json in the workspace is an open pass, and bench/owner says whose. Alive, it is running in another session: collect.py exits 3 naming the holder; say who holds it and stop. Dead, /wikidoc:tri resumes it (route, apply, learn adopt the bench); collect never overwrites it.
  3. Order. The files of the inboxes are read (②), routed (③), decided (④) and applied (⑤) before anything else is touched. Nothing else is in this pass, so nothing waits on it.
  4. Close. Learn (⑥) runs as usual, in its subagent. What an inbox still holds afterwards — an unanswered or refused file, a file left undecided — is named in wiki/state.md.

Done when: every file the inboxes held at collect is filed or named in wiki/state.md, bench/ is archived, and the pass <pass> commit is made.

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.