contextlisted

# Related Context Target (may be empty): $ARGUMENTS Every ledger read goes through one read-only script. The hop files call it `q.sh`; each Bash call is a fresh shell, so always run it by its full path, never through a variable: `${CLAUDE_PLUGIN_ROOT}/skills/context/scripts/q.sh`. Also the answer to "what do I still owe?" — hop 3 alone, filtered, is the whole of it. **Two depths.** The *user* asked "where are we / what's next / continue" with no area named → **hop 0**: run `q.sh summary` and open nothing else. Report in ≤ 10 lines: per area `done/open` and the first open task with its levels · debt `queue` vs `blocked` · last memo. A `next:` whose `Needs` is unticked is not next — say which task it waits on. Stop there. Anything else — a target (area, row, debt `id`, file), a "why is X like this?" question, or **you triggered this yourself before a task** (then the target is that task's area; an empty `$ARGUMENTS` is not a reason for hop 0) → hops 1–4 below, then open **only** the docs those records point at and answer with `file:line` quotes. A follow-up question in the same session digs from where the last hop stopped; never re-run hop 0, never fall back to reading `.claude/clio/docs/` whole. Nothing on record → say so; do not reconstruct an answer from the code. Three things decide whether a change is correct: what the customer asked for (`.claude/clio/docs/specs/`), what was already built and why (`.claude/clio/docs/tasks/`, `.claude/clio/docs/decisions/`), and what