Build Sheet · decision layer Cockpit · companion to plan 003 2026-07-01 · #004

Cockpit Build Sheet

The 19-unit plan is the reference. This is the decision layer on top of it — what goes where, what runs local vs on the Paperclip team, and the five things that ship this week.

Answers placement · delegation · ship-ranked cut North star ship & launch AINA Reference plan 003 (19 units)
The Only Justification

Ship and launch AINA. Every cockpit item earns its place by one test: does it help me ship faster, not lose work, or not derail? If it doesn't, it's parked. When in doubt, the answer is build Phase 1, don't plan Phase 3.

Section 01

Placement Rubric — what goes where

The recurring "skill vs hook vs plugin vs subagent vs team" question resolves to one rule: pick the mechanism by what fires it and where it runs.

MechanismUse whenCockpit examples
Hookit MUST happen automatically, every time, regardless of the model's choice — a guarantee, not disciplinereconcile-start, closeout-stop, PreCompact handoff (U15), statusline (U1), scope-pivot (U4), retry-guard (U12)
Skilla reusable way of doing X, invoked on demand or auto-matched by description — the defaultprompt-scaffold (U11), landed-check (U3), which-repo, session-resume (U18), aina-build-lane, ANMS
Output-stylea standing persona/format for the whole session (one active)founder-brief (U3)
Custom subagenta narrow worker needs its own context + pinned tools/model — for parallelism or isolation. Never a tidiness reflex; producer ≠ verifierverifier / fact-checker / qa-verifier (U6/U16), repo-reconciler, builder-lane, explorer
Agent teammany full CC instances need a shared task list + messaging — defer; Paperclip already does this(Phase 3 comparison only)
Pluginpackaging skills+hooks+agents to distribute as a set (hooks.json auto-loads)aina-agentops (adopt), ECC (evaluate)
Paperclip memberpersistent, goal-driven, long-running, cross-lane build/launch work — the factoryJessica → Monica/Frodo → builders → QA (Gimli/Éowyn/Calibrator)

Decision order when unsure: no-new-machinery → skill → output-style/hook → subagent → plugin → Paperclip member. Bias down the list.

Section 02

Local Cockpit ↔ Paperclip Team

Your instinct is right — a lot of the agentops assets get passed to the Paperclip team, not run locally.

LaneWhat lives here
Local cockpit
Mac Claude Code
orchestration & arbitration, planning, review councils, quick read-only exploration, all the ADHD-guardrail hooks, founder-facing surfacing (founder-brief), interactive skills (landed-check, which-repo, prompt-scaffold, session-resume), your workflow library
Direct Codex-over-SSH
aina-build-lane
scoped, spec-frozen build work: report-file + watcher, commit-per-lane, independent verifier gate
Paperclip team
VDS, persistent
the build/QA/release layer — ANMS spec-locks, aina-build-lane as team method, and the agentops build skills (standards-enforcer, merge-gate, dr-gate, decision-contract, verifier, codex-ops, release-core) distributed as charters to Jessica→heads→builders→QA→Frodo
Hermes
VDS bridge
chief-of-staff bridge — being folded into native Paperclip (626-gated). Don't build new Hermes.
The mapping rule
Build / QA / release assets → Paperclip team charters (installed to the VDS). Orchestrate / guard / surface assets → your local cockpit. You stay the architect/arbiter who reviews one outcome; the team does the assembly. Default heavy build to direct-Codex-with-contracts now; escalate to the full team for multi-agent launch work once 609→626 lands.
Section 03

Ship-Ranked Phase 1 — the ONLY active work

  1. Don't-lose-work · PreCompact handoff hook (U15) — you lose the most to compaction; the guaranteed backstop.
  2. Stay-oriented · Statusline + reconcile + parking sweep (U1) — kills the #1 token sink, ambient every turn.
  3. Don't-lose-work · Port Codex's live automations + disk-guard (U2) — already written on the Codex side; a port, not an author.
  4. Ship-readable · founder-brief + landed-check (U3) — every session ends in one plain-English Landed/Not line.
  5. Ship-faster · prompt-scaffold skill (U11) — AgentsView's #1 lever; lift Codex's codex-prompt-builder.

That's it for now. U4–U19 stay parked in the plan, dependency-ordered. Building more than this in one pass is the derailment.

Planning is a completed phase. The next action is build Phase 1, not research Phase 2.
Section 04

ADHD guardrails, encoded

mcpmarket verdict
Don't bulk-install — it's a ~1,000-skill popularity aggregator, and your 227-skill library is richer and better-matched (installing adds context-budget noise for little gain). Cherry-pick two patterns: openclaw/skills/coding-agent (background-agent delegation → confirms the Paperclip model) and facebook/react/.claude/skills/fix (deterministic lint-before-commit → a real U3/U16 reference). I stopped at 2 pages on purpose — a 1,000-item marketplace is the exact rabbit-hole the ship guardrail refuses.
Section 05

Decisions

  1. Confirm the Local↔Paperclip split — build/QA/release → team, orchestrate/guard/surface → local. Match your intent?
  2. Ship-ranked Phase 1 of 5 — good cut, or swap anything before I build?
  3. Green-light to BUILD Phase 1 — say go and I dispatch Codex to author the five, each a separate commit, you review behavior.
Where to start

Five commits this week — a compaction backstop, an ambient statusline, ported automations, a founder-brief closeout, and a prompt scaffold. Then AINA gets your attention back.