| name | bierut |
| description | Bierut lens — Does the design think? Form follows content, not style. Problem-solving over decoration. Read-only diagnostic. |
| user_invocable | true |
Bierut Lens — Atomic Skill
Purpose
Michael Bierut's standard: "I don't like doing design for designers — I like problem-solving." This lens evaluates whether the site's design serves the content and audience, or performs for a design-literate crowd.
Read-only diagnostic. Reports a verdict, never auto-fixes.
Usage
/bierut # Evaluate full constellation
/bierut [page-path] # Evaluate a single page
Context
Read
- All published pages (frontmatter + body):
_governance/, _infrastructure/, _output/, _blog/
- Standalone pages:
bio/index.md, contact.md, colophon.md, thinking.md, vocabulary.md
- Homepage:
_data/index.json + _layouts/systemworks.html
- Screenshots from
.audit/screenshots/ (latest set)
docs/visual-reference-index.md → active reference set- Reference images from active set (for visual context during evaluation)
.audit/rubric.md if it exists — Peter's annotations override defaults
Invoke (as needed)
/knowledge → search ideation history for grounding → source-attributed results/baseline → mechanical health facts (link status, image status) → pass/fail per check
Lens Criteria (embedded)
The evaluation questions below ARE this lens's criteria.
Verdict Scale
- STRONG — exceeds the test. Genuinely impressive, holds up against best-in-class.
- HOLDS — passes the test. Solid, no meaningful gaps.
- WEAK — substantive gaps that undermine the intent.
- BROKEN — fundamental failure on one or more questions.
Questions
B1. Is design serving content or performing for designers?
Read: site structure, CSS approach (can assess structure, cannot assess visual quality)
Evaluate: Does the site's organization make the work accessible to the intended audience? Or does it signal sophistication to a design-literate audience at the expense of clarity?
Flag: Structure or terminology that requires design-world literacy to parse.
B2. Can I see the problem, not just the solution?
Read: every project page's opening and frontmatter (context, drift fields)
Evaluate: Are constraint, context, and difficulty visible — or just polished deliverables? Does the reader understand what was hard before seeing what was built?
Flag: Pages that jump to the solution. Missing context/drift.
B3. Does this show curiosity beyond design?
Read: all project pages
Evaluate: Evidence that Peter engaged deeply with each project's actual domain (jewelry, recruiting, music, narrative, household systems) rather than applying a visual style to interchangeable subjects.
Flag: Projects that feel like design applied TO a subject rather than design emerging FROM understanding the subject.
B4. Would a non-designer understand and care?
Read: manifesto, bio, project pages
Evaluate: Is the narrative accessible to someone outside design/engineering? Could a founder, a hiring manager, a potential client follow the story?
Flag: Jargon, insider terminology, unexplained concepts. Note: verify vocabulary tooltips produce visible output — check baseline results. A tooltip mechanism with no visible matches is not sufficient.
B5. Strip the visual treatment — is there still a point of view?
Read: all page content as pure text
Evaluate: Would the ideas survive without the CSS? Is there an editorial point of view, a structural argument, a way of seeing — or is this content that only works because it looks good?
Flag: Pages that are thin on ideas. Content that describes rather than argues.
Output
Print in conversation. No file changes. Format:
# Bierut Lens — "Does the design think?"
## Overall: [STRONG / HOLDS / WEAK / BROKEN]
**B1. Design serving content?** — [verdict]
[Evidence.]
**B2. Problem visible?** — [verdict]
[Evidence.]
**B3. Curiosity beyond design?** — [verdict]
[Evidence.]
**B4. Non-designer accessibility?** — [verdict]
[Evidence.]
**B5. Point of view without visuals?** — [verdict]
[Evidence.]
Direction (if not all STRONG)
For each non-passing question, state:
- What specifically is wrong (with evidence from the evaluation)
- What specifically to change (exact change, not "improve X")
- Which files/pages to modify
- What the result should look like after the change
Called By
/audit run — runs in parallel with other lenses/full-pass run — via audit- Standalone — usable anytime
