| name | brand-identity-audit |
| description | Audit a brand's full identity (logo, typography, color, voice, visual language) for consistency, distinctiveness, and gap. Use when the user asks for a brand audit, identity review, brand health check, or wants to evaluate how a brand shows up across surfaces. |
brand-identity-audit
audit a brand identity across every surface it shows up on. find drift, find gaps, find what's distinctive and what's interchangeable.
what this skill does
inventories every visible expression of the brand. measures consistency. finds points of drift. names what's distinctive vs generic.
when to use
- a brand has been running for 12+ months without a structured review
- after a redesign that touched multiple surfaces
- before a launch that puts the brand in front of new audiences
- when leadership says "the brand feels inconsistent" but can't name where
inputs required
before running the audit, collect:
- the canonical brand assets: logo files, type specimens, color tokens, voice guide (if one exists)
- the live surfaces: homepage, app, marketing site, social, ads, email templates, support docs, sales decks, recruiting pages
- the audiences each surface targets: brand surfaces serve different audiences and may legitimately vary
if the canonical assets don't exist yet (or are scattered), say so. an audit needs a baseline.
the audit
inspect each axis. capture distinct values and any drift.
1. logo
- what variants exist (full, mark, monochrome, inverted)?
- which surfaces use which variants?
- minimum size violations?
- clear-space violations?
- positioning consistency (always top-left? sometimes centered?)
2. typography
- which typefaces are in use? compare canonical vs actual.
- type scale: does it match the system? off-scale uses?
- weight usage: is hierarchy consistent (display = bold, body = regular)?
- line-height patterns?
- letter-spacing patterns?
3. color
- which hex values appear? group near-duplicates.
- which surfaces use which palette subset?
- accent color usage: consistent or random?
- contrast: any AA / AAA failures?
4. spacing and layout
- grid adherence
- section spacing patterns
- container widths
5. imagery and illustration
- photography style (lit, color-graded, candid, etc.)
- illustration system (if any)
- iconography (one set or many?)
- ai-generated imagery (worth a separate review pass)
6. voice
- run brand-voice-extraction on samples from each surface
- compare extracted voice against canonical guide (if one exists)
- name surfaces where voice is strongest, weakest, most off
7. motion (if applicable)
- transition durations
- easing curves
- where motion appears vs where it doesn't
8. naming and terminology
- product feature names: consistent across surfaces?
- terminology variants ("dashboard" vs "home" vs "overview"?)
- internal vs external terminology bleeding through?
output shape
brand identity audit — [brand name]
executive summary
- distinctive (the 3-5 things this brand owns)
- generic (the 3-5 things any competitor could say or show)
- drift severity: low / medium / high
- top 5 fixes
surface-by-surface
[for each surface]
- logo: [score / notes]
- typography: [score / notes]
- color: [score / notes]
- voice: [score / notes]
- off-canonical findings
axis-by-axis drift
[for each axis]
- canonical: [the reference]
- actual: [what's in the wild]
- severity: [low / med / high]
- affected surfaces: [list]
- proposed resolution
recommended actions
- immediate (must fix before next ship): [list]
- quarterly cleanup (worth scheduling): [list]
- strategic (requires a decision, not just execution): [list]
scoring rubric
per axis, score 1-5:
- 5: canonical and consistent across every surface
- 4: canonical with isolated drift (single surface)
- 3: drift across multiple surfaces, no clear pattern
- 2: more drift than consistency
- 1: no canonical version evident; every surface is its own brand
rules this skill follows
- don't judge what's distinctive without comparison; pull at least 2 competitors
- "feels off" is not a finding; name the specific axis and surface
- voice drift is often the worst kind because no one notices until someone outside does
- consistency for its own sake isn't the goal; coherence in service of recognition is
what this skill doesn't do
- redesign the brand (that's downstream of this)
- write a new voice guide (use brand-voice-extraction)
- judge taste (it measures coherence; aesthetics are out of scope)
- prescribe colors or fonts (it surfaces drift; design decisions happen separately)
related
- brand-voice-extraction (voice audit, more focused)
- design-system-audit (technical token / component drift, separate concern)
- messaging-architecture (verbal positioning, separate axis)
