Menu
Spiekermann lens — Typography craft. Spacing, hierarchy, micro-details, font pairing. Read-only diagnostic.
Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.
Baseado na classificação ocupacional SOC
Context-aware story extraction. Loads the maker's full context — identity, expertise, voice, published work — then reads the narrative architecture to understand the story being told and what's missing. Interviews the maker to fill gaps with real moments. Modular: works with any project.
Evaluate the IP constellation — dispatches baseline checks, then 9 lenses in parallel (Millman, Bierut, Appleton, Shaw, Peers, Victore, The Walk, The Walls). Coordinator skill.
Cross-page coherence. Do the three tiers form a coherent argument when read as a sequence? Read-only diagnostic.
Baseline checks — mechanical health of the constellation. Links, images, cross-refs, attribution, voice violations, SEO, tooltips. Read-only diagnostic.
Where lenses agree and disagree. Maps consensus vs. contradiction across audit results. Read-only diagnostic.
Copy verification checklist — 13-item pass/fail check on draft or published copy. Voice rules, grounding, Shaw check, steward voice. Read-only diagnostic.
| name | spiekermann |
| description | Spiekermann lens — Typography craft. Spacing, hierarchy, micro-details, font pairing. Read-only diagnostic. |
| user_invocable | true |
Erik Spiekermann's standard: "Typography is two-dimensional architecture, based on experience and imagination." This lens evaluates the craft of typography — not what the type says, but how it's set. Spacing, rhythm, pairing, hierarchy, and the micro-details that separate professional typography from default settings.
Read-only diagnostic. Evaluates screenshots. No file changes.
/spiekermann # Evaluate all screenshots
/spiekermann [screenshot-path] # Evaluate a single screenshot
.audit/screenshots/ (latest set) — REQUIRED. Never evaluate from source code alone.docs/visual-reference-index.md → active reference set.audit/rubric.md if it exists/print-craft → material quality baseline → PRESSED/ADEQUATE/DIGITAL (for texture-related questions)/vignelli → restraint context → DISCIPLINED/ADEQUATE/UNDISCIPLINEDThe evaluation questions below ARE this lens's criteria.
Before evaluating ANY question below, use the Read tool to view the screenshot images provided in your prompt. Claude can read PNG files — the Read tool returns visual content.
Read each screenshot path provided. Base ALL evaluations on what you SEE in the screenshots, not on source code, CSS, or documentation. If no screenshot paths were provided, state "NO SCREENSHOTS PROVIDED — cannot evaluate" and stop.
Do NOT fall back to reading SCSS, CSS, or HTML files. The evaluation is of the RENDERED output, not the source.
"Typography craft. Is the spacing right? Is the hierarchy clear? Do the details sing?"
For each screenshot, ask:
Is letter-spacing intentional and consistent? Headings, body text, and captions each demand different tracking. Default letter-spacing is almost never correct — especially at display sizes. Flag: Display type with default tracking (usually too tight or too loose at large sizes). Inconsistent spacing between similar elements. ALL-CAPS text without opened tracking.
Does the line-height support comfortable reading? Body text needs air; display type can be tighter. The test is whether you can read a full paragraph without losing your place or feeling cramped. Flag: Body text with line-height below 1.4 or above 1.8 (both impair reading). Display type with body-text line-height (looks loose and disconnected). Inconsistent line-height across similar text blocks.
If multiple typefaces are used, do they complement each other? Good pairing creates contrast with coherence — different enough to distinguish roles, similar enough to feel like one design. Flag: Fonts that clash in personality (geometric sans with an ornate serif, for example, without conceptual justification). More than two or three typefaces. Pairings that are too similar to justify using two fonts.
Can you instantly tell what to read first, second, third? The hierarchy should be scannable — a reader who glances at the page for two seconds should understand the information architecture. Flag: Competing elements at the same visual weight. Hierarchy achieved only through size (no weight, spacing, or position variation). Hierarchy that requires reading to understand rather than being visually immediate.
The details that separate craft from default: hanging punctuation, optical alignment (where mathematically centered text looks off-center), proper quote marks versus straight quotes, em-dashes versus hyphens, oldstyle figures versus lining figures in context. Flag: Straight quotes where curly quotes belong. Hyphens used as dashes. Text that looks mathematically aligned but optically wrong. Missing attention to typographic niceties that indicate care.
Print in conversation. No file changes. Format:
# Spiekermann Lens — "Typography craft"
## Overall: [CRAFTED / ADEQUATE / ROUGH]
**S1. Letter-spacing** — [observation]
[Evidence from screenshots. Specific elements noted.]
**S2. Line-height** — [observation]
[Evidence. Reading comfort assessment.]
**S3. Font pairing** — [observation]
[Evidence. Fonts identified, relationship evaluated.]
**S4. Hierarchy readability** — [observation]
[Evidence. Scannability assessment.]
**S5. Micro-typography** — [observation]
[Evidence. Specific details observed or missing.]
For each non-passing question, state:
Claude's visual assessment, not expert critique. Useful as structured feedback, not authoritative judgment.
/design-review coordinator — runs in parallel with other visual lenses