Skip to content

Run agency-grade design critique loop #14

Description

@iainherd

Goal

Create a repeatable critique loop that prevents generic AI-generated visual output.

Source artefacts

  • Lir_Website_Design_Language_Spec_v0.2.md §9
  • Lir_Photography_Art_Direction_Brief_v0.1.md
  • User instruction: output should feel like a high-end design/brand agency such as Pentagram.

Acceptance criteria

  • Each major visual slice gets screenshot-based critique.
  • Critique dimensions include composition, typographic tension, originality, restraint, institutional credibility, and “AI-template smell”.
  • At least one design-director review and one browser QA pass happen before sign-off.
  • Iteration notes are stored in repo docs or issue comments.
  • Screenshots are captured as evidence.

Metadata

Metadata

Assignees

No one assigned

    Labels

    agent:brand-strategistBrand strategist: positioning, proof model, strategic fit, Lir brand/archetype integrity.agent:browser-qa-reviewerBrowser QA reviewer: screenshots, responsive, console, accessibility, reduced-motion, Lighthouse/axeagent:design-directorDesign director: owns agency-grade visual critique, anti-AI-template review, design-system judgementagent:orchestratorMorwenna/orchestrator: owns issue sequencing, repo state, PR/deploy verification, and final evidencepriority:highpriority hightype:designtype designtype:qatype qa

    Type

    No type

    Fields

    No fields configured for issues without a type.

    Projects

    No projects

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions