OpenParable
Skip to content
← Back to skills

Architectural Decisions

Testing & Quality
architectural-decisions

Architectural Decisions is an Agent Skill for testing & quality published in the kubernetes-sigs/kueue repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Review a diff for small-scale structural problems — unjustified complexity, duplicated logic, scope creep, or misplaced code.

View on GitHub ↗
QUICK START

How to use this skill

Bring this guide into your coding agent with a prompt tailored to the tool you use.

  1. Open your project in Codex.
  2. Copy and paste the prompt below.
  3. Review files and risks before approving any action.
Prompt to paste
Inspect the Agent Skill at https://github.com/kubernetes-sigs/kueue/blob/HEAD/cmd/experimental/skills/reviewer/architectural-decisions/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/architectural-decisions, but wait for my approval before writing files or running scripts.

Copying this prompt does not install or run the skill. Review third-party files before use.

Skill: Architectural Decisions

Use this when reviewing a diff for small-scale structural problems: how code is organized, how parts connect, and whether complexity is justified. Complex problems may need complex solutions, but every unit of complexity must earn its keep.

Rules

Each rule in this domain is its own sub-skill. Scan the diff for violations of every rule below.

RuleTriggers on
illogical-structurecode a future maintainer will struggle to follow, modify, or extend
nonsensical-decisionsunnecessary indirection, mismatched abstractions, confusing data flow
avoidable-complexitysolutions more elaborate than the problem requires
pointless-intermediate-variablesredundant local variables that add noise without clarity
duplicated-logicidentical blocks across types/adapters/call sites that should be shared — or new helpers with one caller
scope-creepdiffs bundling a bugfix with an unrelated refactor, or over-generalizing a change
misplaced-logiccode placed somewhere a reader would not expect to find it

@./illogical-structure/SKILL.md @./nonsensical-decisions/SKILL.md @./avoidable-complexity/SKILL.md @./pointless-intermediate-variables/SKILL.md @./duplicated-logic/SKILL.md @./scope-creep/SKILL.md @./misplaced-logic/SKILL.md

What not to include

Do not report personal preferences or architectural taste. This skill is for structural problems that materially hurt maintainability, simplicity, or evolution of the code.

In particular, do not flag:

  • a design choice simply because you would have organized the code differently;
  • small helpers, wrappers, locals, or indirections that are reasonable in context, even if you would personally inline or restructure them;
  • broad redesign ideas unless the current structure creates a concrete maintenance risk, ambiguous ownership, duplicated change burden, or extension hazard.

How to report

For each finding, cite the exact file and line, name which rule it violates, and classify severity (high / medium / low). Severity scales with how much the issue hurts maintainability, simplicity, or backward compatibility. Decisions that make the code materially harder to extend are high; cosmetic structural nits are low.

Return two sections:

Findings

A bullet list, one per violation, in this format:

- <High/Medium/Low> | <Finding Title>: <one-sentence explanation grounded in the diff>

Recommendations

One recommendation per finding above, numbered sequentially within this skill. Each recommendation MUST contain all four sections below.

### Recommendation N: <short title>

**Problem**: Describe the specific issue in the diff.
**Reason**: Explain why this is a problem — what goes wrong or degrades over time.
**Solution**: Give a concrete, actionable fix. Where possible, show a before/after code snippet.
**Locations**: List every place this issue occurs, one per line, in `filepath:line` form.