Skill Wiki
Heuristic Review
A structured heuristic evaluation of a UI artifact against Nielsen's 10 usability heuristics. Produces a scored report with evidence-backed issues ranked by severity.
$ prime install @community/method-heuristic-reviewProjection
Always in _index.xml · the agent never has to ask for this.
HeuristicReview [method] v1.0.0
A structured heuristic evaluation of a UI artifact against Nielsen's 10 usability heuristics. Produces a scored report with evidence-backed issues ranked by severity.
Loaded when retrieval picks the atom as adjacent / supporting.
HeuristicReview [method] v1.0.0
A structured heuristic evaluation of a UI artifact against Nielsen's 10 usability heuristics. Produces a scored report with evidence-backed issues ranked by severity.
Input
- Artifact: URL | FilePath | Base64Image
- Depth: "quick" | "dee
Output
- Issues: HeuristicIssue[]
- Scores: HeuristicScores
- Overall: 0..4
- Summary: string
Uses
- @community/step-load-artifact
- @community/step-apply-heuristics
- @community/step-generate-report
Body
- @community/step-load-artifact(artifact) -> document
- @community/step-apply-heuristics(document, @nielsen/taxonomy-10-heuristics) -> scores
- if depth == "deep":
- for each heuristic H where scores.scores[H] < 3:
- deepen analysis of H — read H.chunks/full.md
- gather additional evidence from document.components
- RANK issues by severity (0 = critical first)
- SUMMARIZE: "X/10 heuristics scored good or better. Top issues: [...
- @community/step-generate-report(scores, issues) -> report
Success Criteria
- @community/check-all-heuristics-evaluated
Loaded when retrieval picks the atom as a focal / direct hit.
HeuristicReview [method] v1.0.0
A structured heuristic evaluation of a UI artifact against Nielsen's 10 usability heuristics. Produces a scored report with evidence-backed issues ranked by severity.
Input
- Artifact: URL | FilePath | Base64Image
- Depth: "quick" | "dee
Output
- Issues: HeuristicIssue[]
- Scores: HeuristicScores
- Overall: 0..4
- Summary: string
Uses
- @community/step-load-artifact
- @community/step-apply-heuristics
- @community/step-generate-report
Body
- @community/step-load-artifact(artifact) -> document
- @community/step-apply-heuristics(document, @nielsen/taxonomy-10-heuristics) -> scores
- if depth == "deep":
- for each heuristic H where scores.scores[H] < 3:
- deepen analysis of H — read H.chunks/full.md
- gather additional evidence from document.components
- RANK issues by severity (0 = critical first)
- SUMMARIZE: "X/10 heuristics scored good or better. Top issues: [...
- @community/step-generate-report(scores, issues) -> report
Success Criteria
- @community/check-all-heuristics-evaluated
Input
- Artifact: URL | FilePath | Base64Image
- Depth: "quick" | "dee
Output
- Issues: HeuristicIssue[]
- Scores: HeuristicScores
- Overall: 0..4
- Summary: string
Uses
- @community/step-load-artifact
- @community/step-apply-heuristics
- @community/step-generate-report
Body
- @community/step-load-artifact(artifact) -> document
- @community/step-apply-heuristics(document, @nielsen/taxonomy-10-heuristics) -> scores
- if depth == "deep":
- for each heuristic H where scores.scores[H] < 3:
- deepen analysis of H — read H.chunks/full.md
- gather additional evidence from document.components
- RANK issues by severity (0 = critical first)
- SUMMARIZE: "X/10 heuristics scored good or better. Top issues: [...
- @community/step-generate-report(scores, issues) -> report
Success Criteria
- @community/check-all-heuristics-evaluated
Source
prime-system/examples/frontend-design/primes/compiled/@community/method-heuristic-review/atom.yaml