Domain knowledge skill for Precept brand identity. Covers where brand research lives, how to capture new research, how to use it in decisions, and how to keep it current.
This skill governs how agents work with the brand research corpus — reading it before making decisions, capturing new findings consistently, and keeping it current as the brand evolves.
Applies to: Any agent doing brand-related work — positioning, voice, narrative, visual identity, README copy, hero content, philosophy-adjacent changes, or reviewing brand artifacts. Primary users: J. Peterman (Brand/DevRel), Elaine (UX — when touching brand-system boundary), Frank (boundary decisions), Steinbrenner (positioning overlap).
| Category | Path | Contents |
|---|---|---|
| Brand decisions | design/brand/brand-decisions.md | Canonical decided elements — each marked ✅ (decided) or 🔲 (open) |
| Brand research | research/brand/ | 19 files: positioning, voice, color, typography, visual language, aesthetics, adjacency, philosophy extension, narrative |
| Brand reviews |
design/brand/reviews/ |
| Review artifacts from prior brand work (palette, diagram colors, README) |
| Hero brief | design/brand/hero-creative-brief.md | Creative direction for hero content |
| Brand philosophy | design/brand/philosophy.md | Redirects to docs/philosophy.md |
| Product philosophy | docs/philosophy.md | Grounding document — read-only, never edit without owner approval |
| Brand spec | design/brand/brand-spec.html | Canonical brand artifact (Elaine executes, Peterman supplies meaning) |
| Research index | research/brand/README.md | What goes in research vs. references vs. reviews |
| Explorations | design/brand/explorations/ | HTML prototypes: color, typography, visual language |
| Topic | File |
|---|---|
| Positioning | research/brand/brand-positioning.md, research/brand/adjacent-products.md |
| Voice & tone | research/brand/voice-and-tone.md |
| Narrative | research/brand/brand-narrative.md |
| Color | research/brand/color-systems.md |
| Typography | research/brand/typography.md |
| Visual language | research/brand/visual-language.md |
| Aesthetic precedents | research/brand/aesthetic-brands.md |
| Philosophy | research/brand/philosophy-brand-extension.md, research/brand/philosophy-draft-v2.md, research/brand/philosophy-rewrite-brand-impact.md |
| Data vs state | research/brand/data-vs-state-philosophy.md, research/brand/data-vs-state-architecture.md, research/brand/data-vs-state-runtime.md |
| Brand spec structure | research/brand/brand-spec-structure-research.md |
| README | research/brand/readme-research-peterman.md, research/brand/readme-research-elaine.md |
brand-decisions.md for whether the element is already decided (✅) — cite the existing decisiondocs/philosophy.md when the work touches product identity or positioning| Acceptable | NOT acceptable |
|---|---|
| "Per brand-decisions.md §Voice: 'Authoritative with warmth,' models: Serilog + MediatR + Workflow Core" | "The brand voice is professional" |
| "Per voice-and-tone.md: Formal↔Casual set to 'slightly casual, contractions fine'" | "Keep it conversational" |
| "Per brand-positioning.md: Temporal coined 'durable execution' — same category-creation pattern" | "Category creation is a good strategy" |
| "Per brand-decisions.md §Voice Do/Don't: 'Precept compiles business rules' NOT 'supercharges your logic'" | "Avoid hype language" |
Rule: If a claim could be made by any brand strategist without reading the file, it is not a citation.
When your output includes claims not grounded in the research corpus, flag them:
### Claims from general expertise
- {claim} — not found in brand research; based on {source/reasoning}
| Type of finding | Location |
|---|---|
| Positioning study, voice analysis, visual identity precedent | research/brand/{topic}.md |
| Critique of a specific brand artifact | design/brand/reviews/{artifact}-review-{agent}.md |
| HTML prototype / exploration | design/brand/explorations/{name}.html |
| Approved rule or decision | Update design/brand/brand-decisions.md |
Every research file should include:
research/ — those go in a references/ folderbrand-decisions.md needs a new 🔲 item or an update to an existing onebrand-decisions.md to mark it ✅ with the rationale