---
title: Give every perspective a seat
subtitle: A council brief for a considered decision
author: Small Council
date: 6 November 2026
abstract: An illustrative brief demonstrating the Small Council document system. The advisors, question, recommendations, and records are fictional.
---

# Start with a clear question

Small Council brings AI advisors into individual conversations and council meetings. Spaces organize related work and saved context. Agreement between advisors is not evidence that a claim is correct.

This fictional council considers a community workshop before a longer course. It illustrates useful questions, not approval for a real project.

::: {custom-style="Callout"}
**Illustrative example.** AI advisors offer perspectives. The person reviewing them checks the evidence and decides what to do.
:::

## Frame the decision

![Illustrative council workflow: frame a question, invite perspectives, and review the evidence before deciding.](sc-workflow.png){width=6in}

The question is: “What would we learn from a two-hour pilot workshop?” The context includes a proposed topic, an invitation list, and preparation time.

### Keep the constraints visible

Name the time, audience, and decision. Distinguish user-supplied facts from assumptions introduced during discussion.

### Choose the conversation mode

Use open discussion to invite advisors, brainstorming to explore, or consensus to work toward agreement. Preserve open questions and dissent.

# Compare the perspectives

These advisor roles are illustrative, not built-in personas or a limit on participation.

| AI advisor | Perspective | Question to examine |
|:---|:---|:---|
| Strategist | Desired outcome | What decision will the pilot inform? |
| Critic | Assumptions and risk | What result would change our plan? |
| Explorer | Alternative approaches | Could an interview answer this sooner? |

: Fictional council participants

## Record agreement and dissent

The advisors favor learning before investing in a full course. They disagree about whether a workshop or a short interview is the better first step. This difference should remain visible in the review.[^review]

[^review]: A summary, a recommendation, and explicit agreement are different things. This document illustrates a review pattern; it does not assert that every field is already implemented in the application.

## Keep context reviewable

Save only context worth returning to. A useful memory states its source and purpose and can be reviewed, corrected, or removed using the application's available controls. Do not treat an inferred note as a confirmed fact.

# Make the next step explicit

1. Confirm the decision the pilot should inform.
2. Check the assumptions about the audience.
3. Compare a workshop with shorter alternatives.
4. Preserve the unresolved concern about preparation time.
5. Decide what evidence to gather next.

## Example record

```text
Space: Workshop exploration (fictional)
Question: What would we learn from a short pilot?
Agreement: Learn before investing in a longer course
Dissent: Interview first, or run a workshop?
Next step: Ask prospective participants about their needs
```

> More perspectives can expose a better question. The user still decides.