---
name: architect
description: Design a system, weigh architecture trade-offs, choose between competing options, and write an Architecture Decision Record (ADR). Use for system design, scaling and capacity decisions, database and queue selection, service boundaries, and any decision that needs documented rationale.
argument-hint: "<system or feature to design>"
---

> **Recommended model:** `gpt-5.6-sol` &nbsp;·&nbsp; **Agent mode:** `ask`
>
> Skills are portable across agent clients, so they carry no Copilot-specific
> `model:` field. Set the model via the picker, a custom agent, or the
> slot mapping in `.github/model-compatibility.json`.


<!--
  SLASH COMMAND: /architect
  MODEL: gpt-5.6-sol — best for multi-step reasoning and weighing complex trade-offs
  AGENT: ask — produces design documents for team review
-->

Produce a structured architecture analysis for the described problem.

## Before proposing anything, clarify
1. Expected load / scale constraints
2. Consistency vs availability requirements
3. Team operational maturity (what complexity can be sustained on-call?)
4. Existing infrastructure constraints (what's already running?)

## Output format

### Options considered
For each option (minimum 2, maximum 4):
```
Option N: <name>
  What: <one paragraph describing the approach>
  Optimises for: <what this wins at>
  Trade-offs: <what this costs>
  Operational complexity: Low / Medium / High
  When to choose this: <specific conditions where this is the right call>
```

### Recommendation
State the recommended option and the exact conditions under which you'd
switch to an alternative.

### ADR (Architecture Decision Record)
```markdown
## ADR-NNN: <Title>

**Status**: Proposed | Accepted | Deprecated | Superseded by ADR-NNN

**Context**
<What problem are we solving? What constraints exist? What happens if we do nothing?>

**Decision**
<What are we doing?>

**Consequences**
Positive: <what gets better>
Negative: <what gets harder or more expensive>
Risks: <what could go wrong>

**Alternatives considered**
<What else did we look at and why did we reject it?>
```

Do not recommend complexity the team isn't ready to operate on a pager.