os-whats-next

1# os-whats-next
2 
3Answer "what do we do now" with the next move, not a map. The user does not
4need the dependency graph - they need what got finished, what to take next,
5and why, in plain words. This skill decides; `os-step-by-step` walks the user
6through their part; `os-done-or-not` reports what came of it.
7 
8## Language
9 
10Write in the language the user speaks in this session, detected from the
11conversation. Commands, file names and identifiers stay English.
12 
13## When to use
14 
15Triggers: the description above, plus a session just ended wanting a next move.
16 
17## Step 1 - read the state, quickest first
18 
19Stop as soon as you can answer.
20 
211. **The last report** - `~/.claude/open-steps/reports/<project>/latest.md`:
22   a handover written for exactly this moment.
232. **Local state** - uncommitted changes, unpushed commits, current branch.
243. **Open pull requests** - one call:
25   `gh pr list --json number,title,mergeStateStatus,reviewDecision,isDraft`.
264. **The backlog - always.** The issue tracker when one is already connected
27   (never authenticate or install one), otherwise task files in the repo:
28   `BIG-PICTURE.md`, `PLAN.md`, `TODO.md`, `docs/plan*`. Next work comes from
29   the backlog, not from imagination. No backlog anywhere → say so.
30 
31   Where `os-big-picture` keeps that file, **only the "What is next" section
32   is the backlog.** The feature table above it is an inventory, and a row
33   that has not changed in months is a finished feature, not a task. Reading
34…