ship-status

Shipyard Status Dashboard

Read all project state, validate it, auto-fix issues, and present a clear dashboard.

Context

!shipyard-context diagnose !shipyard-context status-dashboard

Paths. All file ops use the absolute SHIPYARD_DATA prefix from the context block. No ~, $HOME, or shell variables in file_path. Bash is for shipyard-context (reads), the onboarding command surfaced by context, and shipyard-data lock status|release plus shipyard-data metrics regenerate (CLI-owned housekeeping — see Check 6/7) ONLY — no other shipyard-data subcommand and no other shell. Never use echo/printf/shell redirects to write state files — use the Write tool (auto-approved for SHIPYARD_DATA) for the reconcile-log and sentinel files this skill maintains. Metrics are CLI-owned: memory/metrics.json is authoritative and memory/metrics.md is generated; never Write/Edit either metrics file. /ship-status consumes pipeline cursor, PROGRESS.md, and SPRINT.md summaries from shipyard-context status-dashboard; it never writes them and never parses cursor frontmatter itself (the PreToolUse hook denies model writes to those; the shipyard-data/shipyard-context CLIs are their only state interface). HANDOFF.md is retired — a paused pipeline is a cursor with status: paused. The two skill-mutex lock files (.active-session.json, .active-execution.json) are CLI-owned too — this skill never hand-Writes them, only shipyard-data lock status (read) / lock release ... --force (clear).

Onboarding gate. If the bundled context contains SHIPYARD_ONBOARDING_REQUIRED=true, run the exact SHIPYARD_ONBOARDING_COMMAND once with Bash, report the CLI output to the user, and STOP. Do not infer setup state by reading or writing Shipyard state files; onboarding decisions are CLI-owned.

Render before asking. Before every AskUserQuestion, render the decision context — the scenarios, concrete examples, tradeoffs, and any verbatim content being approved — as chat text; the tool call then carries only the short question and option labels. A bare AskUserQuestion with no rendered context above it is a bug (the window is too small to carry a real decision). Content that exists only in a Read result, a subagent/Agent return, a dossier file, or the question/option strings themselves does not count as rendered (the UI shows a compact card) — restate it as assistant chat text immediately above the ask.