Skip to content
Back to skills

Architect

ASecurity

Design types, interfaces, and module boundaries for an explicit architecture request or an unresolved consequential design choice.

  • 2 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added November 7, 2026
development

Security analysis

A100/100

Pro scans all 6 files and shows the line behind each finding

Scanned November 7, 2026

npx -y skills add BumpyClock/agent-templates --skill architect --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Architect?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Architect
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/bumpyclock-architect-agent-templates/badge)](https://www.skillsdirectory.com/skills/bumpyclock-architect-agent-templates)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: architect
description: Design types, interfaces, and module boundaries for an explicit architecture request or an unresolved consequential design choice.
disable-model-invocation: true
---

# Architect

Resolve the design question before implementation.
A mechanical caller or signature change does not need a design exercise.

## Scope

Inspect the relevant contracts and consumers.
Use [How](../how/SKILL.md) when an architectural explanation would resolve a material uncertainty.
Reuse an existing source model instead of repeating subsystem exploration.

Sketch only the types, signatures, or module boundaries needed to assess the decision.
Compare distinct designs when the user requests alternatives or consequential uncertainty remains.
Use independent candidates when separate expertise justifies the cost and delegation is available.
Do not require a fixed candidate count or an unavailable comparison skill.
For a requested multi-candidate review, use [the candidate brief](references/subagent-prompt.md) and [design rationale](references/rationale-template.md).

Assess interface depth, ownership, and consumer requirements.
Use [design red flags](references/design-red-flags.md) when a proposed abstraction needs review.
Use [architecture guidance](../../programming/references/architecture/architecture-planning.md) for prerequisite and dependency decisions.

## Completion

For design-only requests, return the recommendation, necessary sketch, tradeoffs, and unresolved decisions.
For authorized implementation, continue through [Programming](../../programming/SKILL.md) without another approval gate.
Pause when the user requests a checkpoint or a missing decision materially changes scope.

If implementation disproves the design, reassess the affected assumption.
Revise the sketch or compare another design when evidence warrants it.
Do not restart the complete investigation for an ordinary parameter correction.

Use [verification guidance](../../programming/references/verification-before-completion.md) to assess the result and reuse valid evidence.

Files in this skill

  • SKILL.md2 KB
  • agents/openai.yaml272 B
  • references/design-red-flags.md1.9 KB
  • references/rationale-template.md3 KB
  • references/runner-prompt.md2.9 KB
  • references/subagent-prompt.md3 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.