architecture-critique

Evaluate technical software plans against an 11-dimension architectural rubric.

Updated Jun 12, 2026
One-click install
npx skills add https://github.com/ambrovia/agent-skills-pipeline --skill architecture-critique
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: architecture-critique
Source: https://github.com/ambrovia/agent-skills-pipeline/tree/main/skills/architecture-critique
Command: npx skills add https://github.com/ambrovia/agent-skills-pipeline --skill architecture-critique

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill solves the issue of biased self-evaluation in software planning by enforcing a strict producer-evaluator separation, ensuring technical plans are objectively audited before any code is written.

Core Features & Use Cases

  • Multi-dimensional Scoring: Evaluates plans across 11 architectural dimensions including spec alignment, layer integrity, and failure-mode coverage.
  • Adversarial Review: Uses a dedicated pipeline-reviewer persona to challenge assumptions and identify implementability holes.
  • Use Case: After a developer generates a technical plan for a new feature, run this skill to receive a scored critique and actionable feedback, preventing scope creep and architectural drift before implementation begins.

Quick Start

Invoke the architecture-critique skill to audit the technical plan for the current work package and generate a scoring report.

Frequently Asked Questions about architecture-critique

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I audit a technical software plan for architectural rigor before implementation?

Audit a technical software plan by evaluating it against an 11-dimension architectural rubric to ensure feasibility and spec alignment. This process applies during the design phase to challenge assumptions and identify implementability holes before code is written.

What is the best way to prevent architectural drift and scope creep in software design?

Prevent architectural drift by enforcing a strict producer-evaluator separation using an adversarial pipeline-reviewer persona. This objective critique generates a scored report and actionable feedback, stopping scope creep before implementation begins.

Can I use an automated architectural review to check layer integrity and failure-mode coverage?

Yes, an automated architectural review evaluates plans across 11 dimensions, explicitly including layer integrity and failure-mode coverage. It audits technical software plans against defined architectural invariants to ensure spec alignment.

How does adversarial code review improve software design quality assurance?

Adversarial code review improves software design quality assurance by using a dedicated pipeline-reviewer persona to challenge assumptions and identify implementability holes. This eliminates biased self-evaluation and enforces objective plan auditing.

Do I need defined architectural invariants to run a software plan critique?

Yes, running a software plan critique requires adherence to defined architectural invariants and project-specific configuration rules. These constraints provide the baseline needed to evaluate technical plans against the 11-dimension rubric accurately.

When should I run an architecture critique during the software development workflow?

Run an architecture critique during the design phase of the software development workflow, after generating a technical plan but before writing code. This timing prevents architectural drift by ensuring feasibility and spec alignment early.