---
name: task-content-review
type: task
description: Monthly Mr Pritchard capture review — sweeps the parked captures the weekly qualify pass set aside and asks whether they now cluster, by theme, into something shareable that no single one was on its own. Promotes an emergent theme as a synthesising piece for Aled; leaves the rest parked for next month. Proposes only; never bins a capture. Linear only; editorial-only. Self-contained task.
license: Proprietary — Aled Pritchard workspace use.
---

# task-content-review

The **monthly aggregation pass** of the Mr Pritchard content loop. The weekly `task-content-drain` qualifies fresh captures and parks the ones that aren't publishable on their own. This task is the second look: once a month it reads the parked pile and asks a *different* question — not "is this one worth publishing?" but **"do these now add up to something?"** A single capture can be too small to publish while five of them, seen together, are a clear argument. Linear connector only. Follows `convention-linear` and, for anything it proposes to write, `convention-mrp-voice`.

**Editorial-only**, Mr Pritchard team only. **Proposes only; never bins a capture** — the whole point of parking is that nothing is thrown away.

## Behaviour

### Pass 1 — Read the parked pile

Read the **parked captures**: `capture`-labelled issues still in **Backlog** in the Articles project (i.e. not promoted to Defined by a weekly pass). Group them by their **`theme:*`** label.

### Pass 2 — Aggregate by theme

For each theme cluster, ask: **do these captures now form something shareable that no single one was on its own?** Look for a through-line — a repeated conviction, a pattern across moments, an argument that only appears at volume. Be honest: most months, most clusters won't have cohered yet, and "not yet" is the right answer.

### Pass 3 — Promote or leave parked

- **Cohered** — the cluster is now an argument worth publishing. **Propose a synthesising piece**: a comment on a new `PIECE (idea)` issue (or the strongest capture in the cluster), assigned to Aled, stating the through-line and **linking the constituent captures** as related issues. On Aled's pick, the weekly `task-content-drain` develops and drafts it against `convention-mrp-voice`. The captures stay linked as its raw material.
- **Not yet** — leave the cluster parked. It rolls into next month's review. No capture is closed, binned, or downgraded.

### Pass 4 — Report

Short summary: parked captures reviewed, clusters by theme, any emergent piece proposed (with the captures it draws on), what stays parked. No preamble, no sign-off.

## Where it sits

The two cadences together are the whole funnel: **`task-content-drain`** (weekly) qualifies the fresh and parks the rest; **this task** (monthly) aggregates the parked for emergent themes. A capture's only exits are *published* (via a piece) or *still parked* — never deleted. Voice, when anything is drafted, is always `convention-mrp-voice`.

## Guardrails

- **Never bins a capture.** Parking is permanent until a capture is published or the operator removes it by hand.
- **Aggregates, doesn't re-qualify.** The weekly pass already judged each capture alone; this pass judges them *together*, by theme.
- **Proposes only** — it proposes a synthesising piece; Aled picks; the weekly drain drafts. It writes no piece itself.
- **Honest about incoherence** — "these haven't cohered yet" is a valid and common outcome; don't force a theme.
- **Editorial-only** — Mr Pritchard team only. Linear only — never code, the delivery loop, or Pipeline.
- Aled is the gate.

## Setup

- Deployed as a **Cowork scheduled task** (cloud), **monthly**. Connector: Linear only.
- Self-contained (no separate skill).
- Output: emergent-piece proposals in Linear (not a file).

Owned by Reach (R3-ACH); the voice is Mr Pritchard editorial canon.

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