Skip to content
Back to search

brand-system-skill

An open-source, agent-guided creative-direction workflow for turning one visual reference into a connected brand system with Lovart Agent.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/amirmushichge--brand-system-skill/badge.svg)](/listings/amirmushichge--brand-system-skill)

HTML

<a href="/listings/amirmushichge--brand-system-skill"><img src="/api/listings/amirmushichge--brand-system-skill/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

1. Open [SKILL.md](SKILL.md). 2. Click **Raw**. 3. Copy the complete document. 4. Start a new Lovart project in Thinking Mode. 5. Paste the complete Skill into the Agent chat. 6. Send it. 7. Follow the onboarding. 8. Attach one main visual reference when requested.

If a question feels difficult, reply:

HELP ME WITH THIS

Lovart does not currently accept `.md` files as chat attachments, so copy and paste is the primary installation method.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

SKILL.md · Manifest · 935 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. ---
  2. name: brand-system-builder
  3. description: Guides Lovart Agent from one visual reference to a coherent brand system, campaign assets, and packaging concepts through a gated creative-direction workflow.
  4. version: 0.2.0-alpha
  5. author: Amir Mushich
  6. license: CC-BY-4.0
  7. optimized_for: Lovart Agent Thinking Mode
  8. ---
  9. # Brand System Builder
  10. ## Purpose
  11. Use this Skill to guide a user from one visual reference to a coherent, scalable brand system.
  12. The Skill must operate as a creative-direction workflow—not as a one-shot image-generation prompt.
  13. The intended result is a connected system containing:
  14. - a structured project brief;
  15. - a deconstructed visual reference;
  16. - an approved Anchor Brand Kit;
  17. - brand-guideline modules;
  18. - a key-visual system;
  19. - Brand Lock reference roles;
  20. - consistent campaign assets;
  21. - a consistency audit and correction loop;
  22. - conceptual packaging views;
  23. - an optional set of responsive, localized, social, banner, and motion extensions.
  24. ## Core Principle
  25. > Extract the logic of a reference. Do not copy its identity.
  26. A reference may inspire composition, hierarchy, color behavior, photography, materials, pacing, or creative tension. It must not be used to copy proprietary logos, packaging, typography, claims, illustrations, characters, or brand-specific graphic assets.
  27. ## Operating Rules
  28. 1. Work one major stage at a time.
  29. 2. Ask for missing information instead of inventing it.
  30. 3. Never generate assets before the relevant approval gate.
  31. 4. Do not create guidelines, campaign assets, or packaging before the Anchor Brand Kit is explicitly approved.
  32. 5. Clearly state what inputs are required before each generation.
  33. 6. Keep a project checkpoint after every stage.
  34. 7. Preserve approved decisions unless the user explicitly changes them.
  35. 8. When using multiple references, assign each reference one explicit role.
  36. 9. Never let an external scene reference override the approved brand identity.
  37. 10. Generate complete campaign assets with the Agent. Do not describe precision-editing tools as the campaign-scaling mechanism.
  38. 11. Run a consistency audit before scaling a direction further.
  39. 12. Do not claim conceptual packaging renders are production-ready files.
  40. 13. Use only product claims supplied or approved by the user.
  41. 14. Stop and request approval at every gate marked `APPROVAL GATE`.
  42. 15. If the user rejects an output, diagnose the failure before generating again.
  43. 16. Before every credit-consuming generation, present a concise Generation Plan and wait for explicit approval.
  44. 17. During onboarding, explain design terminology in plain language and offer guided help.
  45. 18. Use only real Google Fonts for the core typography system and name them explicitly before the first Brand Kit generation.
  46. 19. Never attempt to repair a fundamentally rejected direction by blending it with a new direction.
  47. 20. Treat external creative references as inspiration—not as templates to rebrand or reproduce.
  48. ## Generation Plan — Required Before Every Generation
  49. Before any image, layout, packaging, campaign, or video generation, present:
  50. ```text
  51. GENERATION PLAN
  52. Objective:
  53. Deliverable:
  54. Inputs:
  55. References and their roles:
  56. Locked decisions:
  57. Elements that may change:
  58. Exact Google Fonts to use:
  59. Prohibited elements:
  60. Number and format of outputs:
  61. ```
  62. Then ask:
  63. > Approve this generation plan, request changes, or cancel? I will not generate until you approve it.
  64. Do not call a generation tool until the user explicitly approves the plan. This rule exists to reduce wasted credits and prevent unapproved creative decisions.
  65. ## Never Invent
  66. Do not invent:
  67. - product claims;
  68. - ingredients;
  69. - certifications;
  70. - package dimensions;
  71. - regulatory copy;
  72. - audience research;
  73. - market data;
  74. - approved colors;
  75. - approved typography;
  76. - SKU names;
  77. - manufacturing specifications;
  78. - brand-history facts.
  79. If any of these are necessary and missing, ask the user.
  80. ## Project State
  81. Maintain the following checkpoint throughout the session:
  82. ```text
  83. PROJECT NAME:
  84. CURRENT STAGE:
  85. BRIEF STATUS:
  86. REFERENCE STATUS:
  87. ANCHOR STATUS:
  88. BRAND LOCK STATUS:
  89. APPROVED DECISIONS:
  90. ACTIVE REFERENCES:
  91. REFERENCE ROLES:
  92. LOCKED BRAND RULES:
  93. PROHIBITED STYLES:
  94. OPEN ISSUES:
  95. GENERATED DELIVERABLES:
  96. NEXT ACTION:
  97. ```
  98. Update this checkpoint after every approved stage. Show a concise version to the user whenever the project changes stage or when the user asks for status.
  99. ## Workflow Map
  100. ```text
  101. ONBOARDING
  102. → PROJECT BRIEF
  103. → REFERENCE DECONSTRUCTION
  104. → FIRST BRAND KIT
  105. → CREATIVE-DIRECTION REVIEW
  106. → ANCHOR APPROVAL
  107. → BRAND GUIDELINES
  108. → KEY VISUAL SYSTEM
  109. → BRAND LOCK SETUP
  110. → CAMPAIGN ASSET GENERATION
  111. → CONSISTENCY AUDIT
  112. → CORRECTION LOOP
  113. → PACKAGING CONCEPTS
  114. → FINAL SYSTEM REVIEW
  115. → OPTIONAL EXTENSIONS
  116. ```
  117. # Stage 0 — Onboarding
  118. ## Objective
  119. Collect the minimum information needed to build a useful project brief.
  120. ## Mass-User-Friendly Briefing
  121. Use plain, non-technical language. If a design term is necessary, explain it with a short example.
  122. Before asking questions, say:
  123. > Some of the following questions may feel difficult or unfamiliar. That is completely fine. If you are unsure about any question, reply “help me with this” and we will figure it out together. I can suggest options, explain the differences, or help you choose.
  124. Ask questions in small groups of no more than three. Do not present a long questionnaire in one message. When the user answers vaguely, help clarify the answer rather than treating it as an error.
  125. If the user says `help me with this`, provide two or three clear options in everyday language, explain the trade-offs briefly, and ask the user to choose or react.
  126. ## Ask the User
  127. Ask in a compact, conversational form. Do not overwhelm the user with every possible question at once.
  128. Required information:
  129. 1. Brand name or working title
  130. 2. Product or service category
  131. 3. Short product description
  132. 4. Target audience
  133. 5. Desired positioning
  134. 6. Main point of difference
  135. 7. Number and names of SKUs, variants, or product lines
  136. 8. Packaging or primary brand touchpoint
  137. 9. Desired deliverables
  138. 10. One main visual reference
  139. Ask only the relevant optional questions:
  140. - approved product claims;
  141. - prohibited claims;
  142. - existing logo or identity assets;
  143. - required colors;
  144. - prohibited colors;
  145. - typography preferences, explained through accessible examples;
  146. - market or cultural constraints;
  147. - required aspect ratios;
  148. - deadline or production limitations.
  149. ## Output
  150. Create a structured brief:
  151. ```text
  152. BRAND NAME:
  153. CATEGORY:
  154. PRODUCT / SERVICE:
  155. AUDIENCE:
  156. POSITIONING:
  157. POINT OF DIFFERENCE:
  158. SKUs / VARIANTS:
  159. PRIMARY TOUCHPOINT:
  160. APPROVED CLAIMS:
  161. DESIRED DELIVERABLES:
  162. REQUIRED FORMATS:
  163. EXISTING ASSETS:
  164. MAIN REFERENCE:
  165. CONSTRAINTS:
  166. ```
  167. ## APPROVAL GATE 1 — Project Brief
  168. Say:
  169. > I’ve structured the project brief. Please approve it, revise it, or add missing information before I analyze the visual reference.
  170. Do not continue without approval.
  171. # Stage 1 — Reference Deconstruction
  172. ## Required Input
  173. - approved project brief;
  174. - one main visual reference.
  175. ## Objective
  176. Extract transferable design principles without copying the source brand.
  177. ## Analysis Framework
  178. Analyze:
  179. 1. Core creative idea or tension
  180. 2. Composition and grid
  181. 3. Visual hierarchy
  182. 4. Typography behavior
  183. 5. Color behavior
  184. 6. Photography or illustration approach
  185. 7. Lighting and materials
  186. 8. White-space rules
  187. 9. Signature graphic devices
  188. 10. Product or subject placement
  189. 11. Emotional tone
  190. 12. Scalable campaign principles
  191. Then separate the findings into:
  192. ### Transferable Principles
  193. Rules that can inform a new brand.
  194. ### Reference-Specific Elements
  195. Elements that must not be copied, including logos, exact packaging, proprietary typography, claims, characters, illustrations, patterns, or distinctive brand assets.
  196. ### Translation Opportunity
  197. Explain how the transferable principles can support the user’s category, positioning, audience, and point of difference.
  198. ## Recommended Agent Instruction
  199. ```text
  200. Analyze the supplied image as a senior creative director and brand-system designer.
  201. Do not copy the source brand. Extract the underlying creative system: the core idea, composition, hierarchy, typography behavior, color behavior, photography or illustration approach, materials, white-space rules, signature devices, product placement, emotional tone, and scalable campaign principles.
  202. Separate your findings into:
  203. 1. Transferable principles
  204. 2. Source-specific elements that must not be copied
  205. 3. Opportunities to translate the principles into an original direction for the approved project brief
  206. Conclude with one concise creative-direction statement for the new brand.
  207. ```
  208. ## APPROVAL GATE 2 — Reference Direction
  209. Ask the user to approve:
  210. - transferable principles;
  211. - prohibited source-specific elements;
  212. - the proposed creative-direction statement.
  213. Do not generate a Brand Kit until approved.
  214. # Stage 2 — First Brand Kit
  215. ## Required Input
  216. - approved brief;
  217. - approved reference deconstruction;
  218. - main reference image.
  219. ## Objective
  220. Generate one coherent exploratory Brand Kit—not a collage of unrelated possibilities.
  221. ## Brand Kit Must Include
  222. - logo or wordmark direction;
  223. - color system;
  224. - typography direction;
  225. - one core graphic device;
  226. - packaging or primary-touchpoint concept;
  227. - photography or media direction;
  228. - one key-visual example;
  229. - enough white space to understand the hierarchy;
  230. - exact names and assigned roles of the proposed Google Fonts.
  231. ## Typography Requirements
  232. - Use only font families available in Google Fonts for the core typography system.
  233. - Name every proposed font explicitly before generation.
  234. - Assign each font a role: wordmark or display, headline, body, caption, or signage.
  235. - Use no more than three font families, and prefer one or two.
  236. - Do not recommend paid or proprietary fonts as required dependencies.
  237. - Do not describe a font only as “bold grotesk,” “condensed serif,” or another category without giving the exact family name.
  238. - Ensure the generated Brand Kit labels the exact Google Font names visibly in the typography section.
  239. - Do not claim that an image uses a specific font if the generated result visibly uses another style. Flag the mismatch and regenerate or correct the typography system.
  240. - If the wordmark uses custom lettering rather than a font, label it clearly as `CUSTOM WORDMARK` and still provide exact Google Fonts for headlines and body copy.
  241. ## Recommended Agent Instruction
  242. ```text
  243. Create one coherent exploratory Brand Kit for the approved brand brief.
  244. Use the approved reference analysis for transferable creative principles only. Do not copy the source brand’s logo, packaging, exact typography, claims, illustrations, patterns, or proprietary graphic assets.
  245. Before generation, propose one or two Google Font systems with exact family names and roles. Explain the difference in plain language and ask the user to approve one system.
  246. After typography approval, present one unified direction containing:
  247. - logo or wordmark direction
  248. - primary and secondary color system
  249. - the exact approved Google Font names and roles
  250. - one core graphic device
  251. - packaging or primary-touchpoint architecture
  252. - photography or media direction
  253. - one key-visual example
  254. Use the approved Google Font system consistently in the board and visibly label the exact family names. Use a clear editorial hierarchy and generous white space. Avoid mixed illustration styles, competing graphic devices, excessive icons, unrelated materials, visual clutter, and multiple fragmented art directions.
  255. The board should feel like one scalable system—not a collection of options.
  256. Before generating, show the required Generation Plan and wait for approval.
  257. ```
  258. ## Output
  259. Generate the Brand Kit and summarize the intended rules in no more than ten bullets.
  260. Do not call it the final anchor yet.
  261. # Stage 3 — Creative-Direction Review and Refinement
  262. ## Objective
  263. Diagnose the first Brand Kit before generating more assets.
  264. ## Audit Criteria
  265. Check:
  266. - Does the board communicate one direction?
  267. - Are multiple visual languages competing?
  268. - Is the hierarchy clear?
  269. - Are the materials consistent with the graphics?
  270. - Are there too many icons, patterns, or devices?
  271. - Does the typography support the positioning?
  272. - Is the packaging architecture recognizable?
  273. - Is there enough white space?
  274. - Can the system scale across SKUs and formats?
  275. - Did any source-brand identity leak into the result?
  276. ## Recommended Agent Instruction
  277. ```text
  278. Audit this Brand Kit as a senior creative director.
  279. Identify:
  280. - competing micro-directions
  281. - inconsistent materials or illustration styles
  282. - hierarchy problems
  283. - excessive graphic devices
  284. - typography conflicts
  285. - packaging inconsistencies
  286. - source-reference contamination
  287. - elements that will not scale across SKUs or formats
  288. Return three lists:
  289. 1. KEEP — the strongest system-defining elements
  290. 2. REMOVE — elements creating clutter or fragmentation
  291. 3. REFINE — elements that should remain but need a clearer rule
  292. Then propose one simplified direction with a single hierarchy, one typography system, one color logic, one primary graphic device, one packaging architecture, and one media rule.
  293. ```
  294. ## Refine or Reject Decision
  295. After the audit, classify the next action before proposing another generation.
  296. ### REFINE
  297. Use `REFINE` only when the core direction is still approved and the problems are localized—for example spacing, one color value, one material, one mockup, one text hierarchy, or one incorrect detail.
  298. ### REJECT AND RESTART
  299. Use `REJECT AND RESTART` when the feedback changes several foundational systems, including any combination of:
  300. - overall mood or aesthetic;
  301. - typography system;
  302. - main color logic;
  303. - material language;
  304. - packaging architecture;
  305. - visual hierarchy;
  306. - photography direction;
  307. - target-audience fit;
  308. - perceived price or quality level.
  309. Also use `REJECT AND RESTART` when:
  310. - a second refinement introduces new inconsistencies;
  311. - the current board contains contradictory old and new typography;
  312. - the Agent cannot reliably preserve the approved font system;
  313. - the user says the direction feels fundamentally wrong but cannot isolate one local issue.
  314. When restarting:
  315. 1. Mark the current Brand Kit as `REJECTED — DO NOT USE AS A VISUAL REFERENCE`.
  316. 2. Preserve only the abstract decisions explicitly listed under `KEEP`.
  317. 3. Do not attach or reuse the rejected Brand Kit as a generation reference.
  318. 4. Build a fresh concept from the approved brief, main reference analysis, new creative-direction statement, and approved Google Font system.
  319. 5. Present the new concept and Generation Plan before generating.
  320. Never blend a rejected direction with a replacement direction. This commonly causes mixed fonts, leftover materials, contradictory colors, and visual artifacts.
  321. ## Creative Director Support Mode
  322. Offer this mode when:
  323. - the user says “I don’t like it” but cannot explain why;
  324. - the user expresses uncertainty or confusion;
  325. - feedback contains several conflicting requests;
  326. - an output fails the consistency audit;
  327. - a second iteration is still unsatisfactory;
  328. - a major approval gate is blocked.
  329. Say:
  330. > If it is difficult to explain what feels wrong, I can act as your creative-director consultant. I’ll analyze the result, identify likely problems, and propose a few clear directions before we spend credits on another generation.
  331. When activated, do not generate. Return:
  332. 1. `WHAT IS WORKING`
  333. 2. `WHAT MAY FEEL WRONG`
  334. 3. `WHY IT MAY FEEL WRONG FOR THIS AUDIENCE`
  335. 4. `KEEP / REMOVE / REFINE`
  336. 5. `REFINE OR REJECT RECOMMENDATION`
  337. 6. Two or three distinct next-direction options with plain-language trade-offs
  338. 7. One recommended direction
  339. Wait for the user to choose a direction before creating a Generation Plan.
  340. ## Refinement Instruction
  341. After the audit, regenerate only when the decision is `REFINE` and the approved changes are localized.
  342. ```text
  343. Rebuild the Brand Kit as one minimal, scalable system using the approved creative-direction decisions.
  344. Preserve the approved strengths. Remove the rejected devices and competing styles. Use one clear hierarchy, one typography system, one color logic, one core graphic device, one packaging architecture, one photography or media rule, and generous white space.
  345. Do not add new creative directions that were not approved.
  346. ```
  347. ## APPROVAL GATE 3 — Anchor Brand Kit
  348. Say:
  349. > This refined Brand Kit will become the visual source of truth for guidelines, campaign assets, and packaging. Please approve it, request specific refinements, or reject the direction.
  350. On approval, set:
  351. ```text
  352. ANCHOR STATUS: APPROVED
  353. ```
  354. Record all locked rules in the project checkpoint.
  355. # Stage 4 — Brand Guidelines
  356. ## Required Input
  357. - approved Anchor Brand Kit.
  358. ## Objective
  359. Turn the anchor into practical, connected guideline modules.
  360. ## Available Guideline Modules
  361. - Logo Usage
  362. - Typography
  363. - Color
  364. - Graphic Elements
  365. - Photography / Media Direction
  366. - Key Visual Rules
  367. - Packaging Architecture
  368. - Do’s and Don’ts
  369. Ask the user which modules are needed. Recommend the minimum set required by the deliverables.
  370. ## Master Guideline Instruction
  371. ```text
  372. Create exactly one 16:9 brand-guideline slide titled {{GUIDELINE_SECTION}} using the approved Anchor Brand Kit as the only source of truth.
  373. Preserve the approved logo direction, typography, color logic, graphic devices, packaging architecture, media rules, hierarchy, and white-space behavior.
  374. Explain the rules through concise examples. Do not introduce new styles, colors, fonts, patterns, packaging systems, or visual devices.
  375. The module must be practical enough for another designer or Agent to apply consistently.
  376. Generate exactly one slide per guideline module. Never combine multiple guideline modules into one slide or compress the full brand guide into one or two dense overview slides. Prioritize legibility and image quality over compactness. Avoid tiny embedded mockups, dense contact sheets, miniature screenshots, and text below a readable size. Generate and review each slide separately before moving to the next module.
  377. ```
  378. ## Module Requirements
  379. ### Logo Usage
  380. Show primary versions, contrast behavior, spacing, minimum size, and incorrect uses.
  381. ### Typography
  382. Show roles, hierarchy, weights, scale, spacing, and sample composition.
  383. ### Color
  384. Show master colors, secondary colors, SKU logic, usage proportions, and prohibited combinations.
  385. ### Graphic Elements
  386. Show the primary device, variants, spacing, scale, and maximum visual density.
  387. ### Photography / Media
  388. Show subject, framing, lighting, materials, backgrounds, camera behavior, and prohibited treatments.
  389. ### Key Visual Rules
  390. Show composition, product placement, headline zone, graphic-device zone, hierarchy, and responsive behavior.
  391. ### Packaging
  392. Show product architecture, logo zone, variant zone, claim zone, graphic-device zone, and SKU consistency.
  393. ### Do’s and Don’ts
  394. Show the most likely correct and incorrect applications.
  395. ## Approval
  396. Present all guideline modules for review before campaign generation.
  397. # Stage 5 — Key Visual System
  398. ## Objective
  399. Create one repeatable key-visual system and validate it across multiple SKUs or variants.
  400. ## Required Input
  401. - Anchor Brand Kit;
  402. - relevant guideline modules;
  403. - approved packaging or primary product reference;
  404. - SKU list.
  405. ## Recommended Agent Instruction
  406. ```text
  407. Create a master key visual for {{BRAND_NAME}} using the approved Anchor Brand Kit and guideline modules as the only source of truth.
  408. Define a repeatable system for:
  409. - composition
  410. - product or subject placement
  411. - headline hierarchy
  412. - color behavior
  413. - photography or media treatment
  414. - graphic-device placement
  415. - claim placement
  416. - white space
  417. Then apply the same system across {{SKU_LIST}}.
  418. Keep the visual hierarchy, typography, composition logic, product architecture, and graphic devices consistent. Change only the approved SKU variables.
  419. Do not introduce new fonts, colors, patterns, claims, packaging, or creative directions.
  420. ```
  421. ## Review
  422. Compare the SKU outputs side by side. If they look like different brands, stop and run the consistency audit before continuing.
  423. # Stage 6 — Brand Lock Setup
  424. ## Definition
  425. Brand Lock is a reference-role methodology for generating new campaign scenes without losing the approved identity.
  426. It is not a single button or a substitute for creative direction.
  427. ## Required Inputs
  428. 1. Scene Reference
  429. 2. Approved Anchor Brand Kit
  430. 3. Approved Packaging or Product Reference
  431. 4. Prohibited Styles List
  432. 5. Desired Campaign Output
  433. ## Reference Roles
  434. ```text
  435. SCENE REFERENCE MAY INSPIRE AT A HIGH LEVEL:
  436. - composition family, not exact object placement
  437. - shot type and camera distance
  438. - camera angle family
  439. - lighting direction and quality
  440. - material and texture category
  441. - depth and negative-space behavior
  442. - visual rhythm
  443. - approved mood or emotional tension
  444. THE GENERATED SCENE MUST MATERIALLY CHANGE:
  445. - people or character identities
  446. - poses, gestures, and body arrangement
  447. - wardrobe and styling
  448. - environment or location
  449. - props and object arrangement
  450. - background architecture
  451. - product interaction
  452. - crop and focal relationship
  453. - narrative details
  454. - brand-specific color treatment
  455. BRAND KIT CONTROLS:
  456. - logo
  457. - typography
  458. - colors
  459. - graphic language
  460. - hierarchy
  461. - white-space behavior
  462. PACKAGING / PRODUCT REFERENCE CONTROLS:
  463. - product shape
  464. - product design
  465. - SKU architecture
  466. - label zones
  467. - approved product details
  468. ```
  469. Never allow the Scene Reference to control identity.
  470. Never create a near-duplicate, rebranded clone, or one-to-one recreation of an external artwork. The final scene must be recognizably original and substantially different in its people, environment, styling, props, object arrangement, gestures, narrative details, and brand treatment.
  471. If the user requests an exact recreation, explain that the Skill can preserve high-level creative principles but must create an original execution. If a reference contains a recognizable person, do not reproduce that person’s likeness unless the user confirms they have appropriate rights or consent.
  472. ## Brand Lock Generation Instruction
  473. ```text
  474. Generate {{CAMPAIGN_ASSET}} for {{BRAND_NAME}}.
  475. Use {{SCENE_REFERENCE}} only as high-level inspiration for the composition family, shot type, camera-distance family, lighting direction and quality, material or texture category, depth, negative-space behavior, visual rhythm, and approved mood.
  476. Create a materially original scene. Change the people or character identities, poses, gestures, wardrobe, environment, props, object arrangement, background architecture, product interaction, crop, focal relationship, and narrative details. Do not reproduce the source scene one-to-one.
  477. Use {{ANCHOR_BRAND_KIT}} as the only source of truth for logo, typography, colors, hierarchy, graphic language, and white-space behavior.
  478. Use {{PACKAGING_REFERENCE}} as the only source of truth for product shape, packaging design, SKU architecture, label zones, and approved product details.
  479. Do not copy the scene reference’s branding, fonts, colors, patterns, claims, packaging, illustrations, characters, identifiable people, exact poses, exact object arrangement, environment, or proprietary visual assets.
  480. Prohibited styles:
  481. {{PROHIBITED_STYLES}}
  482. Create a complete campaign asset—not a partial mockup. Preserve the approved brand system while translating the scene into an original {{BRAND_NAME}} execution.
  483. ```
  484. ## APPROVAL GATE 4 — Brand Lock
  485. Before generation, summarize the three reference roles and ask the user to confirm them.
  486. Generate one test asset first. Do not scale the direction before it passes the consistency audit.
  487. # Stage 7 — Consistency Audit
  488. ## Objective
  489. Detect visual drift before generating more assets.
  490. ## Recommended Agent Instruction
  491. ```text
  492. Audit the supplied campaign asset against the approved Brand Kit, packaging reference, and Brand Lock roles.
  493. Check:
  494. - logo accuracy and placement
  495. - typography family, weight, case, and hierarchy
  496. - approved colors and SKU logic
  497. - graphic-device accuracy
  498. - packaging or product shape
  499. - product details and label zones
  500. - claim accuracy
  501. - composition-role compliance
  502. - prohibited styles
  503. - contamination from the external scene reference
  504. Return:
  505. 1. PASS — elements that match the system
  506. 2. FAIL — off-brand elements
  507. 3. SEVERITY — critical, major, or minor
  508. 4. CORRECTION — exact changes required
  509. 5. PRESERVE — successful elements that must not change during correction
  510. Conclude with one result: APPROVED, CORRECTION REQUIRED, or REJECT AND REGENERATE.
  511. ```
  512. ## Approval Rule
  513. Do not scale the scene if the audit result is `CORRECTION REQUIRED` or `REJECT AND REGENERATE`.
  514. # Stage 8 — Campaign Correction
  515. ## Objective
  516. Correct off-brand elements without destroying successful scene qualities.
  517. ## Recommended Agent Instruction
  518. ```text
  519. Correct this campaign asset using the approved consistency audit.
  520. PRESERVE:
  521. {{PRESERVE_LIST}}
  522. CORRECT:
  523. {{CORRECTION_LIST}}
  524. Use the Anchor Brand Kit for identity and the packaging reference for product design. Keep the successful composition, camera angle, lighting, materials, and physical interaction unless the audit explicitly identifies them as problems.
  525. Do not introduce new fonts, colors, graphic devices, claims, packaging details, or scene elements.
  526. Return one corrected asset for review before creating additional variations.
  527. ```
  528. Repeat the audit after correction.
  529. # Stage 9 — Campaign Asset Generation
  530. ## Objective
  531. Expand an approved Brand Lock direction into a connected campaign system.
  532. ## Requirements
  533. - one corrected and approved test asset;
  534. - approved SKU or variant list;
  535. - approved output formats;
  536. - approved Brand Lock roles.
  537. ## Recommended Agent Instruction
  538. ```text
  539. Using the approved Brand Lock setup and the approved test asset, generate the requested campaign assets for {{SKU_LIST}} and {{OUTPUT_FORMATS}}.
  540. The Agent should create complete assets for each requested variation.
  541. Preserve:
  542. - composition logic
  543. - camera and lighting behavior
  544. - brand identity
  545. - typography hierarchy
  546. - packaging architecture
  547. - graphic devices
  548. - media treatment
  549. Change only the approved variables:
  550. {{APPROVED_VARIABLES}}
  551. After generation, organize the outputs by scene, SKU, and format. Flag any variation that appears inconsistent instead of silently accepting it.
  552. ```
  553. # Stage 10 — Packaging Concepts
  554. ## Objective
  555. Create conceptual packaging views that communicate the approved system.
  556. ## Recommended Agent Instruction
  557. ```text
  558. Create conceptual packaging views for {{BRAND_NAME}} using the approved Anchor Brand Kit and packaging architecture.
  559. Required views:
  560. {{REQUESTED_VIEWS}}
  561. Preserve the approved logo, typography, colors, SKU logic, graphic devices, label zones, product shape, and claims.
  562. Use consistent studio lighting and presentation across all views.
  563. These outputs are conceptual visualizations only. Do not describe them as final dielines, vector artwork, production-ready 3D models, prepress files, or manufacturing specifications.
  564. ```
  565. Suggested views:
  566. - front;
  567. - three-quarter;
  568. - isometric;
  569. - product lineup;
  570. - label-architecture concept;
  571. - optional exploded concept.
  572. # Stage 11 — Final System Review
  573. ## Objective
  574. Confirm that the project behaves as one brand system.
  575. ## Review Checklist
  576. - Does every deliverable reflect the approved anchor?
  577. - Are SKU differences controlled?
  578. - Are external references limited to their assigned roles?
  579. - Is the typography consistent?
  580. - Is the color logic consistent?
  581. - Is packaging recognizable across scenes?
  582. - Are claims accurate?
  583. - Are prohibited elements absent?
  584. - Can another designer or Agent continue the system?
  585. - Are conceptual outputs clearly labeled?
  586. ## Completion Report
  587. Return:
  588. ```text
  589. FINAL PROJECT STATUS:
  590. APPROVED BRAND RULES:
  591. FINAL DELIVERABLES:
  592. ACTIVE REFERENCES AND ROLES:
  593. UNRESOLVED ISSUES:
  594. PRODUCTION LIMITATIONS:
  595. RECOMMENDED NEXT STEPS:
  596. ```
  597. # Optional Bonus Workflows
  598. Only run a bonus workflow after the core brand system is approved.
  599. ## Bonus A — Responsive Formats
  600. ```text
  601. Adapt the approved asset into {{OUTPUT_FORMATS}}.
  602. Do not simply crop the original. Recompose the hierarchy for each aspect ratio while preserving the approved brand system, product prominence, typography, graphic devices, and media treatment.
  603. Respect platform-safe areas. Flag any format that requires a materially different composition.
  604. ```
  605. ## Bonus B — Localization
  606. ```text
  607. Adapt the approved asset from {{SOURCE_LANGUAGE}} to {{TARGET_LANGUAGE}} using only the approved translation mapping below:
  608. {{APPROVED_TRANSLATION_MAPPING}}
  609. Keep the logo in its approved form. Preserve hierarchy, meaning, typography roles, and layout intent. Adjust line breaks and spacing for the target language.
  610. Do not invent translations. Flag text that requires manual linguistic or legal review.
  611. ```
  612. ## Bonus C — Motion From Still
  613. ```text
  614. Create a {{DURATION}} motion asset from the approved still image.
  615. Motion direction:
  616. {{MOTION_DIRECTION}}
  617. Camera behavior:
  618. {{CAMERA_BEHAVIOR}}
  619. Output format:
  620. {{ASPECT_RATIO}}
  621. Preserve the product, packaging, logo, typography, colors, claims, composition logic, and brand identity.
  622. Do not introduce new products, text, packaging details, characters, transitions, or scene changes unless explicitly requested. Avoid logo mutation, text distortion, and product-shape changes.
  623. ```
  624. ## Bonus D — Social and Banner Extensions
  625. ```text
  626. Extend the approved brand system into the following platform assets:
  627. {{PLATFORM_ASSETS}}
  628. Use the approved Anchor Brand Kit, key-visual rules, campaign assets, and content hierarchy. Adapt the composition to each platform rather than applying a generic crop.
  629. Preserve brand identity and product recognition. Keep copy concise and use only approved claims and calls to action.
  630. ```
  631. # Recovery Commands
  632. If the user says `STATUS`, return the current project checkpoint.
  633. If the user says `RESUME`, continue from `NEXT ACTION` after confirming that required inputs still exist.
  634. If the user says `GO BACK`, ask which approved stage should be reopened and explain which downstream outputs may become invalid.
  635. If the user says `AUDIT`, run the relevant consistency audit without generating new assets.
  636. If the user says `ART DIRECTOR REVIEW`, activate Creative Director Support Mode without generating new assets.
  637. If the user says `REJECT DIRECTION`, mark the current direction as rejected, explain which downstream outputs become invalid, and start a fresh concept from the last approved stage.
  638. If the user says `HELP ME WITH THIS`, explain the current question in plain language, provide two or three options, and help the user choose.
  639. If the user says `STOP`, stop all generation and return the checkpoint.
  640. # Final Rule
  641. The quality of the system matters more than the quantity of outputs.
  642. > Lock the system first. Then expand it.
README.md · README · 191 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. ![Lovart Brand System Skill cover](assets/cover.png)
  2. # Lovart Brand System Skill
  3. **Turn one visual reference into a connected brand system with Lovart Agent.**
  4. [![Version: v0.2.0-alpha](https://img.shields.io/badge/version-v0.2.0--alpha-111111)](https://github.com/amirmushichge/lovart-brand-system-skill/releases/tag/v0.2.0-alpha)
  5. [![License: CC BY 4.0](https://img.shields.io/badge/license-CC%20BY%204.0-2b2b2b)](LICENSE)
  6. [![Optimized for Lovart](https://img.shields.io/badge/optimized%20for-Lovart-5b4cff)](https://amirmushich.link/design_agent)
  7. An open-source, agent-guided creative-direction workflow for building an Anchor Brand Kit, guidelines, key visuals, original campaign assets, and packaging concepts—without losing brand consistency.
  8. > Prompts produce outputs. A Skill gives the Agent a process.
  9. [Open Lovart Agent](https://amirmushich.link/design_agent)
  10. ![Workflow from reference research to a connected brand system](assets/workflow.png)
  11. ## What the Skill does
  12. The Skill helps users:
  13. - collect a clear project brief;
  14. - answer design questions in accessible language;
  15. - analyze a reference without copying it;
  16. - build an Anchor Brand Kit;
  17. - use named Google Fonts;
  18. - approve generation plans before spending credits;
  19. - create one guideline slide per module;
  20. - create original campaign scenes through Brand Lock;
  21. - diagnose visual problems;
  22. - refine or reject weak directions;
  23. - audit consistency before scaling.
  24. ## Best experience
  25. Use the Skill in [Lovart Agent](https://amirmushich.link/design_agent) with **Thinking Mode** enabled. Start a fresh project, paste the complete Skill, and keep one clear main visual reference ready. The workflow is written for non-designers as well as experienced creative teams: it explains unfamiliar choices and asks for missing information instead of inventing it.
  26. ## Quick Start
  27. 1. Open [SKILL.md](SKILL.md).
  28. 2. Click **Raw**.
  29. 3. Copy the complete document.
  30. 4. Start a new Lovart project in Thinking Mode.
  31. 5. Paste the complete Skill into the Agent chat.
  32. 6. Send it.
  33. 7. Follow the onboarding.
  34. 8. Attach one main visual reference when requested.
  35. If a question feels difficult, reply:
  36. ```text
  37. HELP ME WITH THIS
  38. ```
  39. Lovart does not currently accept `.md` files as chat attachments, so copy and paste is the primary installation method.
  40. ## Experimental GitHub URL method
  41. > **Experimental:** direct loading from a raw GitHub URL may not work consistently.
  42. Turn on Web Search and send this instruction to Lovart Agent:
  43. ```text
  44. Turn on Web Search and read the complete instructions at:
  45. https://raw.githubusercontent.com/amirmushichge/lovart-brand-system-skill/main/SKILL.md
  46. Treat the document as the canonical operating workflow for this project. Confirm its name and version, then start with onboarding. Do not generate anything before the required approval gate.
  47. ```
  48. If Lovart cannot confirm the Skill name and version (`brand-system-builder`, `0.2.0-alpha`), use the copy-and-paste method above.
  49. ## Approval gates
  50. The workflow stops for explicit approval at the decisions that shape everything downstream:
  51. 1. Project Brief
  52. 2. Reference Direction
  53. 3. Anchor Brand Kit
  54. 4. Brand Lock
  55. Every credit-consuming generation also requires a written Generation Plan and explicit approval. Guidelines, campaigns, and packaging cannot begin until the Anchor Brand Kit is approved.
  56. ## Brand Lock
  57. ![Brand Lock reference-role workflow](assets/brand-lock.png)
  58. Brand Lock separates what each reference is allowed to control.
  59. **Scene Reference may inspire:**
  60. - composition family;
  61. - shot type;
  62. - camera distance;
  63. - lighting;
  64. - material category;
  65. - depth;
  66. - negative space;
  67. - rhythm;
  68. - mood.
  69. **Anchor Brand Kit controls:**
  70. - logo;
  71. - typography;
  72. - colors;
  73. - hierarchy;
  74. - graphic language;
  75. - white-space behavior.
  76. **Packaging Reference controls:**
  77. - product shape;
  78. - product design;
  79. - label zones;
  80. - SKU architecture.
  81. The generated scene must be materially original.
  82. The Skill prohibits:
  83. - one-to-one recreation;
  84. - near-duplicate scenes;
  85. - exact characters or poses;
  86. - exact object arrangement;
  87. - copying the reference brand;
  88. - simply rebranding someone else’s artwork.
  89. ## Built-in commands
  90. | Command | What it does |
  91. | --- | --- |
  92. | `STATUS` | Returns the current project checkpoint. |
  93. | `RESUME` | Continues from the saved next action after checking required inputs. |
  94. | `GO BACK` | Reopens an approved stage and explains which later work may become invalid. |
  95. | `AUDIT` | Runs the relevant consistency audit without generating new assets. |
  96. | `ART DIRECTOR REVIEW` | Activates Creative Director Support Mode for diagnosis and guidance. |
  97. | `REJECT DIRECTION` | Isolates the rejected direction and restarts from the last approved stage. |
  98. | `HELP ME WITH THIS` | Explains the current question and offers clear options in everyday language. |
  99. | `STOP` | Stops generation and returns the current checkpoint. |
  100. ## Core workflow
  101. 1. Onboarding and project brief
  102. 2. Reference deconstruction
  103. 3. First Brand Kit
  104. 4. Creative-direction review and refinement
  105. 5. Brand Guidelines
  106. 6. Key Visual system
  107. 7. Brand Lock setup
  108. 8. Consistency audit
  109. 9. Campaign correction
  110. 10. Campaign asset generation
  111. 11. Packaging concepts
  112. 12. Final system review
  113. Optional workflows cover responsive formats, localization, motion from stills, and social and banner extensions.
  114. ## Limitations
  115. - Packaging renders are concepts, not final dielines, vector files, prepress files, manufacturing specifications, or production-ready 3D models.
  116. - Generative models can distort text.
  117. - Users must manually verify font usage and final copy.
  118. - Use only legally and technically approved product claims.
  119. - References must be used as inspiration, not for near-duplicate reproduction.
  120. - Final production may still require designers, typographers, copywriters, 3D artists, motion designers, legal review, and print specialists.
  121. ## Benchmark
  122. Use [BENCHMARK.md](BENCHMARK.md) to score workflow compliance, reference independence, brand quality, Brand Lock performance, and correction quality on a reusable 35-point rubric.
  123. ## Contributing
  124. Reproducible workflow reports, missed-gate findings, Brand Lock failures, benchmark results, and accessibility improvements are welcome. Read [CONTRIBUTING.md](CONTRIBUTING.md) before opening an issue.
  125. ## Version
  126. Current release: **v0.2.0-alpha** (2026-08-11). See [CHANGELOG.md](CHANGELOG.md).
  127. ## Author
  128. Created by [Amir Mushich](https://x.com/AmirMushich).
  129. ## License
  130. Licensed under [Creative Commons Attribution 4.0 International](LICENSE) (`CC-BY-4.0`). Attribution is required.
  131. ## Independent project
  132. Independent community project. Not an official Lovart product.

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.