MCPcopy Create free account
hub / github.com/garrytan/gstack / designQualityJudge

Function designQualityJudge

test/skill-e2e.test.ts:1949–1967  ·  view source on GitHub ↗

* LLM judge for DESIGN.md quality — checks font blacklist compliance, * coherence, specificity, and AI slop avoidance.

(designMd: string)

Source from the content-addressed store, hash-verified

1947 * coherence, specificity, and AI slop avoidance.
1948 */
1949async function designQualityJudge(designMd: string): Promise<{ passed: boolean; reasoning: string }> {
1950 return callJudge<{ passed: boolean; reasoning: string }>(`You are evaluating a generated DESIGN.md file for quality.
1951
1952Evaluate against these criteria — ALL must pass for an overall "passed: true":
19531. Does NOT recommend Inter, Roboto, Arial, Helvetica, Open Sans, Lato, Montserrat, or Poppins as primary fonts
19542. Aesthetic direction is coherent with color approach (e.g., brutalist aesthetic doesn't pair with expressive color without explanation)
19553. Font recommendations include specific font names (not generic like "a sans-serif font")
19564. Color palette includes actual hex values, not placeholders like "[hex]"
19575. Rationale is provided for major decisions (not just "because it looks good")
19586. No AI slop patterns: purple gradients mentioned positively, "3-column feature grid" language, generic marketing speak
19597. Product context is reflected in design choices (civic tech → should have appropriate, professional aesthetic)
1960
1961DESIGN.md content:
1962\`\`\`
1963${designMd}
1964\`\`\`
1965
1966Return JSON: { "passed": true/false, "reasoning": "one paragraph explaining your evaluation" }`);
1967}
1968
1969describeIfSelected('Design Consultation E2E', [
1970 'design-consultation-core', 'design-consultation-research',

Callers 1

skill-e2e.test.tsFile · 0.70

Calls 1

callJudgeFunction · 0.90

Tested by

no test coverage detected