okhp3-brand-style-registry
Skill3 starsChanged 2 months agoMITView on GitHub
OverKill Hill P³ visual style registry and application workflow. Use when a user wants to extract a reusable brand, visual style, palette, typography system, layout language, or interface aura from a website, stylesheet, document, presentation, infographic, or set of visual assets. Also activate when a user wants an SPA, documentation, presentation, image brief, or other target artifact aligned to one or more named style profiles. This is the authoritative profile-driven workflow for capturing and applying visual style without blending unrelated brands or treating inferred cues as declared rules.
In this repository57 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .agents/skills/vercel-react-best-practices/agents.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- architecture-decision-recordsSkill
- catalog-integritySkill
- frontend-designSkill
- okhp3-artifact-validationSkill
- okhp3-as-is-process-captureSkill
- okhp3-capture-intakeSkill
- okhp3-chatgpt-project-migrationSkill
- okhp3-custom-gpt-builderSkill
- okhp3-custom-gpt-readinessSkill
- okhp3-decision-model-authoringSkill
- okhp3-elicitation-interviewsSkill
- okhp3-equilibrium-reviewSkill
- okhp3-evidence-standardSkill
- okhp3-future-state-strategySkill
- okhp3-gpt-skill-conversion-planSkill
- okhp3-graduation-gateSkill
- okhp3-handoff-packagingSkill
- okhp3-notion-capture-routerSkill
- okhp3-outcome-modeling-coreSkill
- okhp3-overkill-hill-brandSkill
- okhp3-process-captureSkill
- okhp3-process-gap-analysisSkill
- okhp3-process-intake-and-scopeSkill
- okhp3-project-promotionSkill
- okhp3-replit-build-in-publicSkill
- okhp3-replit-canvas-boardSkill
- okhp3-replit-design-pipelineSkill
- okhp3-replit-free-mode-autonomySkill
- okhp3-replit-github-syncSkill
- okhp3-replit-multi-artifactSkill
- okhp3-replit-repl-janitorSkill
- okhp3-replit-repository-janitorSkill
- okhp3-replit-task-executorSkill
- okhp3-repository-janitorSkill
- okhp3-repository-organizerSkill
- okhp3-session-handoffSkill
- okhp3-skill-catalogerSkill
- okhp3-skill-discoverySkill
- okhp3-skill-foundrySkill
- okhp3-source-backed-researchSkill
- okhp3-thread-context-extractionSkill
- okhp3-thread-extract-chatgptSkill
- okhp3-thread-extract-claudeSkill
- okhp3-triage-and-fileSkill
- okhp3-vite-github-pagesSkill
- production-build-verificationSkill
- published-keyboard-recovery-checkSkill
- publish-health-checkSkill
- publishing-trigger-checkSkill
- static-route-validationSkill
- validation-smokeSkill
- vercel-react-best-practicesSkill
- web-design-guidelinesSkill
- okhp3-skill-promotionSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
18 rules. About 0 are common, 12 are distinctive, and 6 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈12
Distinctive · estimate
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 31
When applying or auditing a profile that carries a brand name:
- Distinctive: line 35
Check the rendered result at narrow container widths (mobile nav, narrow sidebars, compressed columns) specifically, since that is where …
- Distinctive: line 36
If you find an existing broken instance while doing unrelated work in a target repository, flag and fix it as part of that work.
- Distinctive: line 56
Use this layout unless the target repository already has a stronger convention:
- Distinctive: line 65
If the target repository already has a style-library convention, preserve it and map the same profile fields into that convention.
- Very specific: line 69
Read references/style-profile-schema.md before creating or revising a profile.
- Very specific: line 71
When a CSS file is available, run scripts/extract_css_signals.py --input <path> --source <locator> before interpreting it.
- Very specific: line 78
Prefer declared design sources over visual inference.
- Distinctive: line 83
Keep the original source file or a public locator available for review.
- Very specific: line 91
Create one <style-id>.yaml file using the schema.
- Very specific: line 101
Add the profile to registry.yaml with its display name, path, status, and tags.
- Distinctive: line 103
Use a transaction-like update: write the profile to a new or temporary path, validate its identity and required fields, then update the …
- Distinctive: line 109
Before changing a target, show the named profiles available to the user and ask which to use.
- Very specific: line 120
If the user names a specialized wrapper such as okhp3-overkill-hill-brand, treat that wrapper’s bundled profile as the default primary.
- Distinctive: line 125
Update shared primitives before one-off components.
- Distinctive: line 128
Check contrast, focus visibility, font fallbacks, density, and any target-framework constraints.
- Distinctive: line 130
Do not rewrite generated files, dependency lockfiles, or unrelated components merely to make the target look consistent.
- Distinctive: line 173
When the user or verified target repository identifies an AskJamie or Glee-fully workflow, load only its bundled profile before applying …
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
okhp3-brand-style-registry
OverKill Hill P³ · overkillhill.com · github.com/OKHP3
Build a durable library of named visual profiles, then use it to create work that feels intentionally aligned instead of generically themed. Treat a profile as design evidence and reusable direction, not as permission to copy another organization’s identity.
This skill has two operating modes: capture a source into a named profile, or apply one or more existing profiles to a target. Keep those modes separate. A capture updates the library only after provenance and confidence are recorded; an application changes the target only after the profile roles and target boundaries are explicit.
Cardinal rule: brand name integrity
A registered brand name string -- most notably "OverKill Hill P³™" -- must never break, wrap, or truncate across a line boundary in any target artifact (navigation bars, footers, headers, body copy, presentation slides, or exported documents). A mid-name line break (e.g. "OverKill Hill" wrapping away from "P³™") is a rendering defect, not a cosmetic nit, and must be fixed immediately wherever it is found -- including in artifacts this skill did not itself produce.
When applying or auditing a profile that carries a brand name:
- Join the name with non-breaking spaces (
in HTML/JSX,\u00A0elsewhere) so the browser cannot insert a line break between words. - Also set
white-space: nowrap(or the equivalent for the target medium) on the element rendering the name, as a second, independent guard -- the non-breaking spaces protect the name if the element itself can still wrap onto a new line at narrow widths. - Check the rendered result at narrow container widths (mobile nav, narrow sidebars, compressed columns) specifically, since that is where this defect surfaces first.
- If you find an existing broken instance while doing unrelated work in a target repository, flag and fix it as part of that work; do not leave a known brand break unresolved.
This rule applies to every registered brand name a profile declares, not only OverKill Hill P³™.
Scope
| In scope | Out of scope |
| Websites, CSS, visual documents, presentations, infographics, and image assets | Logo recreation, trademark clearance, or identity theft |
| SPA styling, documentation direction, presentation direction, and image briefs | Applying styles with no target owner approval |
| One primary style with carefully bounded supporting styles | Silent palette, type, or layout averaging |
Profile library model
Store each style in its own profile file and list it in a registry. There is no fixed profile limit. The separation preserves provenance, makes profiles independently reusable, and prevents a later capture from overwriting an established identity.
Use this layout unless the target repository already has a stronger convention:
brand-styles/
registry.yaml
profiles/
<style-id>.yamlIf the target repository already has a style-library convention, preserve it and map the same profile fields into that convention. Do not create a second registry merely because the directory name differs.
At capture time, ask the user for a human-readable style name and a stable slug. If the user provides several sources for one identity, confirm that they belong to one profile before combining their evidence. If sources represent different identities, create separate profiles.
Read references/style-profile-schema.md before creating or revising a profile. Use assets/style-library.example.yaml when a starter registry helps.
When a CSS file is available, run scripts/extract_css_signals.py --input <path> --source <locator> before interpreting it. The JSON output is an observation ledger, not a finished profile. Use the same principle for documents, screenshots, and rendered pages: preserve the raw locator or page or region reference before assigning semantic roles.
Phase 1: capture and extract
- Establish source authority. Ask whether each source is owned by the user, publicly referenceable, or only inspirational. Record that answer in the profile.
- Prefer declared design sources over visual inference. CSS variables, token files, design-system documentation, and source artwork are
declaredevidence. Rendered pages and visual artifacts areobservedevidence. Narrative interpretation isinferredevidence. - Extract only reusable cues. Capture color roles, typography roles, spacing, radii, borders, shadows, motion, grid and density, components, imagery direction, icon treatment, and language describing the intended emotional effect.
- Preserve variants. A dark mode, campaign surface, application view, or section theme is a named variant, not a conflicting duplicate token.
- Record uncertainty. Do not turn a visually estimated color or font into a declared token. Use a confidence value and source note.
Keep the original source file or a public locator available for review. When a source cannot be accessed, say so in the profile and continue only with user-supplied material. Never fabricate an extraction from a URL that was not inspected.
For a website, inspect both its declared styling and its rendered hierarchy. For a document or infographic, sample visible visual patterns and mark values as observed unless a style guide establishes them. Do not add private URLs as a required dependency for a public profile.
Phase 2: store
Create one <style-id>.yaml file using the schema. Include at minimum:
- profile identity and use boundary
- source provenance and evidence level
- foundation and semantic color tokens
- typography roles and fallbacks
- layout, surface, interaction, and imagery guidance
- accessible-use notes and target constraints
- variants, conflicts, and confidence notes
Add the profile to registry.yaml with its display name, path, status, and tags. Never replace an existing profile just because its palette shares colors with a new one. Version a changed profile and explain whether it reflects a source update or a corrected extraction.
Use a transaction-like update: write the profile to a new or temporary path, validate its identity and required fields, then update the registry entry. If validation fails, leave the existing profile untouched. Preserve old profiles as archived when a new identity supersedes them.
Phase 3: select and apply
Before changing a target, show the named profiles available to the user and ask which to use. For multiple profiles, require this selection model:
primary: <style-id>
supporting:
- profile: <style-id>
allowed_roles: [imagery, microcopy]The primary profile controls the target’s overall visual grammar: palette hierarchy, typography, surfaces, spacing, component anatomy, and interaction tone. A supporting profile may contribute only the explicitly allowed roles. It cannot silently replace the primary palette, fonts, or layout. If the user wants a true hybrid, create a new named profile with its own evidence and rationale rather than mutating either parent profile.
If the user names a specialized wrapper such as okhp3-overkill-hill-brand, treat that wrapper’s bundled profile as the default primary. Still inspect the target’s existing tokens and report any source conflict before editing.
For an SPA, work in this order:
- Map profile tokens to the application’s token layer or theme variables.
- Update shared primitives before one-off components.
- Apply typography, surfaces, spacing, controls, and imagery direction to the relevant views.
- Inspect the rendered result at representative viewport sizes.
- Check contrast, focus visibility, font fallbacks, density, and any target-framework constraints.
Do not rewrite generated files, dependency lockfiles, or unrelated components merely to make the target look consistent. Prefer a small token-layer change that can be reverted and reviewed.
For documentation, presentations, and image briefs, translate the profile into target-appropriate direction. Preserve hierarchy, tone, proportion, and texture without pretending that CSS can be applied directly to a non-web artifact.
Required handoff
Report four things at the end of a capture or application run:
- The profile or profiles used, including evidence level and confidence.
- The tokens and cues applied or intentionally not applied.
- Any unresolved source conflict, missing asset, or font-licensing concern.
- The target-specific verification completed and the remaining human review decisions.
- Confirmation that every registered brand name renders intact (no mid-name wrap) at the narrowest reviewed viewport -- see the cardinal rule above.
Included seed material
assets/profile-seeds/overkill-hill.yaml- example independent profile from declared OverKill Hill theme sources.assets/profile-seeds/glee-fully.yaml- example independent profile from declared Glee-fully theme sources.assets/profile-seeds/askjamie.yaml- example independent profile from declared AskJamie theme sources.assets/specialization-seeds/- copy-ready narrow skills for the repositories that own those three profiles.
These are examples and specialization seeds, not a universal default visual identity.
References
references/style-profile-schema.md- profile structure, evidence levels, and multi-profile conflict rules.references/source-sampling.md- source-specific sampling, evidence ledger, rights, and confidence rules.references/application-playbook.md- selection, target preflight, application sequence, and handoff contract.assets/style-library.example.yaml- starter registry with separate profile entries.scripts/extract_css_signals.py- deterministic CSS signal extraction for the capture phase.evals/evals.json- three draft evaluation cases for the public workflow.
Optional brand profiles
When the user or verified target repository identifies an AskJamie or Glee-fully workflow, load only its bundled profile before applying this shared procedure:
- AskJamie: workflow profile.
- Glee-fully: workflow profile.
For a neutral or unidentified target, use the shared procedure without a brand profile. An alias or previous package name alone does not establish the target brand. If the requested brand conflicts with the target evidence, resolve that conflict before applying a profile. Do not blend profiles or treat positioning as approved product canon. Profiles cannot weaken evidence, privacy, scope, or authorization requirements. Record the selected profile in the handoff.
Consolidation evidence
Version 1.3.0 bundles the former brand overlays as optional profiles as of 2026-09-19. Earlier evaluation designs and benchmark/results files retain their original versions and bytes as historical records; they do not validate this version or its profile selection. Profile regression cases are defined in the evaluation design and are not live runs. Structural validation is separate from behavioral or performance evidence.
About
Built by Jamie Hill · OverKill Hill P³ Published at github.com/OKHP3 Part of the OKHP3/skillz Agent Skill library. MIT License -- free to use, fork, and adapt. A nod to the source is appreciated.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

