Reviewing the Dashboard Queues

Reviewing the Dashboard Queues

The Queues tab is the heart of your review work. It gathers everything Sera has proposed that is waiting on a human decision, so you never have to hunt through individual databases to find what needs attention.

---

What the Queues tab tracks

The tab surfaces eight review streams:

• KB Drafts: knowledge base articles extracted as drafts
• Canon Changes: proposed governance canon changes, always pending
• Memory Review Queue: institutional memory candidates
• Decisions: decision candidates awaiting confirmation
• Sensitive Review: admin-only sensitive flags
• Transparency Ledger: the CCOS governance ledger entries in draft
• Unowned Tasks: tasks with no owner assigned
• High Risks: risks flagged at high severity

Recommended triage order

Work the queues in this order so the most important and most time-sensitive items are never left waiting:

1. Sensitive Review first. This is admin-only and time-sensitive. Personal, safety, legal, and financial matters cannot wait.
2. High Risks. Anything flagged high severity deserves early eyes.
3. Canon Changes. These shape governance, so review them before routine memory.
4. Memory Queue. Work through the remaining institutional memory candidates.

The weekly goal is simple: get everything in Pending Review down to zero. A clean queue means the system's proposals have all been judged and your memory is current.

---

Understanding the Memory Review Queue

The Memory Review Queue is often the least understood queue, so it is worth being precise about what belongs in it.

It holds the kinds of knowledge that do not fit neatly into a decision, a task, or a risk:

• Relationship context — how people and organizations relate to each other
• Institutional knowledge — how your organization actually works
• Historical facts — what happened and when
• Process insights — lessons about how things get done

Crucially, it does not hold things that are already captured elsewhere. A decision belongs in Decisions, an action belongs in Tasks, and a threat belongs in Risks. The Memory Review Queue is for the connective knowledge that would otherwise be forgotten.

Confidence levels

Each memory candidate carries a confidence level that tells you how strongly Sera believes the underlying content supports it:

• High — a clearly stated fact, expressed directly in the source
• Medium — a reasonable inference drawn from the content
• Low — a passing mention only

Low-confidence items are skipped rather than queued, so what reaches you is worth your attention. High and Medium items are where your judgment adds the most value.

---

The review workflow

Every queued item moves through the same simple lifecycle:

1. Pending Review — waiting for a human
2. Approved — accepted into your memory
3. Rejected — not worth keeping
4. Needs Clarification — held for more information

Approving promotes the candidate into your living record. Rejecting keeps your memory clean. Marking an item Needs Clarification is the honest choice when you cannot yet tell, and it keeps the item visible until someone can resolve it.

---

Making it a habit

Reviewing the queues weekly, in the order above, with the goal of reaching zero Pending Review, is the single most effective habit for keeping your Living Memory Hub trustworthy. The system does the reading and proposing; the queues are where the deciding happens.

One more thing worth knowing: the edits you make while reviewing are not thrown away. A hand-edit made before you click Approve, and a rejection with a reason, are both captured as corrections, and once the same correction has been made consistently, Sera stops making that mistake here. Reviewing is not only quality control; it is how she gets better at this organization specifically.

Key points

The Queues tab tracks KB Drafts, Canon Changes, the Memory Review Queue, Decisions, Sensitive Review, the Transparency Ledger (the CCOS governance ledger), Unowned Tasks, and High Risks. Triage in a deliberate order: Sensitive Review first because it is admin-only and time-sensitive, then High Risks, then Canon Changes, then the Memory Queue, with a weekly goal of clearing all Pending Review to zero. The Memory Review Queue holds relationship context, institutional knowledge, historical facts, and process insights, not items already captured as decisions, tasks, or risks. Confidence levels mark High for a clearly stated fact, Medium for a reasonable inference, and Low for a passing mention that is skipped, and each item moves through Pending Review to Approved, Rejected, or Needs Clarification.