motlin

Restore claude/codex agents from a tmux-reboot snapshot after tmux-resurrect reopens their windows.

Snapshot running tmux claude/codex agents immediately before a reboot so their sessions can be restored afterward.

Commit changes, run git-test precommit checks, and fix failures. Use when asked to fix build, test, lint, typecheck, or precommit failures.

Running precommit checks and build validation. ALWAYS use after ANY code changes.

This skill should be used after completing any task, before returning control to the user. Always run this skill — it handles the case where there's nothing to do.

Coordination guidelines for leading a team of task-executing agents

Workflow guidelines for team members executing tasks

Work with markdown-based task lists in .llm/todo.md files. Use when managing tasks, working with todo lists, extracting incomplete tasks, marking tasks complete, or implementing tasks from a task list.

Render a batch of proposed before/after edits as a self-contained page where the user Accepts/Rejects/Later each card — and optionally applies the accepted edits live. Use when you have concrete, reviewable edit proposals (tag sweeps, text rewrites, bulk node changes) the user should approve before they are applied. Trigger phrases include "review these edits", "let me accept/reject the changes", "proposal review page". NOT for plain status summaries or findings write-ups with no apply step — those belong in the chat, because the Accept/Reject buttons would be inert and misleading.

Advance work-in-progress across many repos by fanning out a subagent tree that rebases every branch, resolves conflicts, tests, and fixes failures — looping while it makes progress and ending with a ✅/❌ tree report. Use when the user wants to advance, process, or sweep their WIP across projects (the manual `j g` → /git:conflicts → `j ta` → /build:fix loop, automated). Accepts optional include/exclude project filters.