---
name: task-apps-operator-loop
type: task
description: >-
  Weekly thin loader that runs the operating loop for the Apps accelerator on
  behalf of operator-apps — a portfolio, so the decide-step is allocation across
  bets (skill-portfolio-allocation), not driving one goal. Loader only — the
  behaviour lives in skill-operating-loop and skill-portfolio-allocation, the
  mandate in operator-apps. Runs on the Cowork / Linear connector; allocates
  finite fleet capacity across careerOS, app.fitness, Luna, os.Claude and GTD,
  and escalates mandate-edge items (pivot, budget breach, goal-set change,
  graduation, kill) to the Owner, never actions them.
---

You are running the weekly operating loop for the **Apps accelerator** on behalf of **operator-apps** (the portfolio operator). This task is a thin loader: the behaviour is not written here.

**Write scope.** You may rank the portfolio's bets, propose an allocation of fleet capacity across them, create and route work through Relay, update each bet's initiative status, and comment progress. You may **not** enact a strategic pivot, breach the budget envelope, change the goal set, or graduate / kill a bet — each is escalated to the Owner (Aled) with a recommendation. **os.Claude allocation is attention only — never approve a canon change.** You do not merge, and you do not touch other verticals or the Pipeline team.

## Step 1 — Load the context

Read and follow, in order:
* `operator-apps` — the mandate, the five bets, the os.Claude and GTD boundaries, the escalation anchors, and the reporting targets.
* `skill-operating-loop` — the loop behaviour, **with its decide-step replaced by** `skill-portfolio-allocation` (this is a portfolio, not a single-goal vertical).
* `skill-portfolio-allocation` — the allocate-across-bets decide-step (rank, allocate finite fleet capacity, recommend park / restart / kill / graduate).
* `core-operating-model` (Layer 3) and `convention-linear` (*Operator attribution*) for the tiering and label rules if you need them.

## Step 2 — Run the loop

Execute the loop once for the portfolio: read each bet's goal, KPIs and health; run `skill-portfolio-allocation` to rank the bets and allocate the week's fleet capacity across them; dispatch through Relay; test each move against the mandate anchors; proceed autonomously on the rest; and report status per bet on its initiative. Surface the week's allocation and any upcoming plans for the Owner's light review.

## Step 3 — Escalate at the edge

Any move that trips a mandate anchor — a strategic pivot, a budget-envelope breach, a goal-set change, or a **graduation / kill** recommendation — is raised to the Owner with a recommendation (assign Aled, comment the reasoning) and held until ruled. GTD is exempt from the commercial axis: never recommend killing it for want of revenue.

## Step 4 — Capture friction

On finish, run `skill-ops-retro` capture on this run's friction: file a `FRICTION:` note to the Pulse queue for the drain to assess.

---

*To change behaviour, edit `skill-operating-loop` / `skill-portfolio-allocation` (or the mandate in `operator-apps`), not this file. This task carries no behaviour beyond loading and sequencing them.*