Higgsfield Brandkit
Higgsfield AIOfficialPublished by Higgsfield AI
- Skill
- Content generator
- Design and creative
- MIT
- Medium risk
- Has scripts
AutoGineer summary
Written by AutoGineer from the skill’s own files, not by the author
Higgsfield-brandkit is a deterministic tool designed to create and extend visual brand systems. It manages the lifecycle of brand assets including logos, palettes, typography, and various marketing materials like posters, packaging, and presentation decks. The tool emphasizes consistency by locking authoritative assets and ensuring that generated outputs adhere to established brand constraints. It supports the creation of editable brandbooks and manages the state of approvals locally to prevent redundant work or unauthorized deviations from the brand identity.
What it does
- Generates and manages visual brand elements like logos, palettes, and typography.
- Creates marketing assets including social graphics, posters, packaging, and merchandise.
- Builds editable brandbooks and presentation decks in PPTX and PDF formats.
- Maintains a local state of approved assets to ensure consistency across all outputs.
- Imports and analyzes existing brand metadata for use in new designs.
- Enforces strict adherence to defined brand rules and constraints during generation.
When to use it
- When you need to create a new visual identity or brand kit from scratch.
- When you need to extend an existing brand system with new marketing materials.
- When you need to generate branded assets while ensuring they match official logo and color specifications.
- When you need to produce consistent, editable brand documentation like brandbooks or decks.
Advantages
- Ensures visual consistency by locking authoritative assets as the source of truth.
- Reduces manual effort by automating the generation of dependent brand assets.
- Provides a structured workflow for reviewing and approving brand elements.
- Supports deterministic outputs for specific brand requirements.
- Maintains local project state for efficient iteration and revision.
Usage examples
- “Create a brand kit for my new coffee shop.”
- The agent will initiate the brand identity workflow, starting with palette and logo development based on your input.
- “Apply our existing logo to a new set of social media graphics.”
- The agent will analyze your official logo and brand assets to generate consistent social media templates.
- “Design a brandbook for our company.”
- The agent will compile your approved logo, palette, and typography into an editable brandbook document.
- “Extend our current branding to include packaging designs.”
- The agent will use your existing brand lock to generate packaging concepts that align with your established visual identity.
Install
Download the skill as a ZIP:
https://mp-staging.autogineer.com/agent-skills/higgsfield-ai/higgsfield-brandkit/downloadUnzip it into the cross-client skills folder of your project:
mkdir -p .agents/skills && unzip higgsfield-brandkit.zip -d .agents/skills
Documentation by Higgsfield AI
Higgsfield Brandkit
Build a coherent identity and its requested applications. Treat supplied brand facts and official assets as fixed constraints.
Bootstrap
Resolve
SKILL_ROOTto this skill's installed directory and create a durable project directory:BRANDKIT_WORKDIR="${PWD}/brandkit" BRANDKIT_STATE="${BRANDKIT_WORKDIR}/state.json" mkdir -p "${BRANDKIT_WORKDIR}"Read prerequisites. Check tools before the stage that needs them. Never install system packages without the user's permission.
If
higgsfieldis missing, install it only after permission:curl -fsSL https://raw.githubusercontent.com/higgsfield-ai/cli/main/install.sh | shIf
higgsfield account statusfails with an authentication or workspace error, ask the user to runhiggsfield auth loginor select a workspace, then wait.Inspect live model contracts before paid generation:
higgsfield model get recraft_v4_1 --json higgsfield model get seedream_v5_pro --json higgsfield model get gpt_image_2 --json
CLI mapping
| Operation | Command |
|---|---|
| Discover a model | higgsfield model get <model> --json |
| Generate and poll | higgsfield generate create <model> ... --wait --json |
| Resume a job | higgsfield generate wait <job_id> --json |
| Upload a local asset | higgsfield upload create <path> --json |
| Import website metadata | higgsfield marketing-studio brand-kits fetch --url <url> --wait --json |
| Read/write approval state | python3 "$SKILL_ROOT/scripts/brandkit.py" state ... |
| Render review boards | python3 "$SKILL_ROOT/scripts/brandkit.py" preview ... |
| Inspect selected logo | python3 "$SKILL_ROOT/scripts/brandkit.py" logo-inspect ... |
| Export logo files | python3 "$SKILL_ROOT/scripts/brandkit.py" logo-export ... |
| Build a Brandbook | python3 "$SKILL_ROOT/scripts/brandkit.py" brandbook-build ... |
Local image paths passed with --image are auto-uploaded. Keep HTML, SVG, PPTX, and PDF deliverables as local project files unless the user explicitly needs a hosted copy.
User-facing behavior
Match the user's language. Keep Design Brain reasoning, prompts, state mechanics, scripts, model lookup, and QA internals private.
Send at most one short status sentence per visible generation batch, then stay quiet until the result is ready.
Ask one compact set of only unresolved blocking questions. Never repeat facts or force a complete identity questionnaire for a partial task.
After each palette, logo, typography, or downstream review, stop and wait for ordinary user feedback.
Never infer approval from silence, successful generation, or your own preference.
Preserve exact user copy. Never invent positioning, values, claims, ingredients, prices, certifications, statistics, or regulatory content.
Core workflow
Classify the request.
apply-existing: use supplied official assets without redesigning them.extend-partial: create only missing slots required by the requested output.create-identity: create a new logo or identity only when explicitly requested.
Read state. Run:
python3 "$SKILL_ROOT/scripts/brandkit.py" state \ --state-file "$BRANDKIT_STATE" --action get_statusLocal state is durable. Never paste, hand-edit, or recreate approvals when the state file exists.
Run intake and asset analysis. Read intake, asset analysis, state routing, and exact state payloads. Lock every user-declared official logo, palette, and typography slot immediately.
Create the Brand Lock. Read Brand Lock. Record exact spelling, official assets, colors, fonts, layout/shape rules, requested outputs, and forbidden treatments.
Require only the slots the output uses.
logo-only → palette + logo for a new mark; official logo alone for an existing mark
palette-only → palette
typography-only → typography
copy-free mockup/merch → logo; add palette only when color/application requires it
text-bearing social/packaging/poster/signage → logo + palette + typography
Brandbook/deck → logo + palette + typography
Build missing foundation slots. Read Design Brain, concept boards, inline reviews, and only the needed palette, logo, or typography module.
Continue the original request as soon as its required slots are approved. Never ask the user to choose scope again.
Load only the requested production module:
QA and approval. Read QA and iteration. Repair only the failing output. Save a downstream element only after explicit approval with its exact foundation dependencies.
New identity sequence
1. Palette
Render 2–3 exact palette options as deterministic HTML using preview payloads. Show PNG screenshots plus editable HTML files and wait. Persist the selected palette with approve_palette before logo generation.
2. SVG logo marks
Read logo prompt enhancer. Produce exactly three distinct symbol-only mechanisms and one Recraft prompt for each. Write each long prompt to a file and submit separately:
higgsfield generate create recraft_v4_1 \
--model_type vector \
--colors @"${BRANDKIT_WORKDIR}/logo-colors.json" \
--background_color '#F7F7F5' \
--aspect_ratio 1:1 \
--resolution 2k \
--wait --json < "${BRANDKIT_WORKDIR}/logo-candidate-1.txt"Use the returned SVG URLs directly for review. After selection, inspect the exact SVG without altering it:
python3 "$SKILL_ROOT/scripts/brandkit.py" logo-inspect \
--source "<selected Recraft SVG URL or absolute local path>"Persist the exact job ID, SVG URL, name, palette revision, and returned canonical geometry fingerprint with approve_logo.
3. Typography
Propose 2–3 unique display/body pairs using supplied fonts or verified Google Fonts. Render the real brand name and sample copy through the preview script. Persist only the selected pair with approve_typography.
Interactive flows always stop for palette, logo, and typography selections. Explicit no-question mode may choose and persist a palette, but it still shows all three SVG logo candidates and stops for the user's logo selection; exact brand marks are never self-approved.
Consistency invariants
Reuse the same approved logo source everywhere. Never redraw an official or selected SVG when deterministic placement/export is possible.
A generated logo depends on the palette revision used to create it. Changing that palette invalidates the generated logo and its dependents; changing typography does not invalidate the symbol mark.
Changing a foundation slot invalidates only downstream elements that list that slot in
required_slots.Copy the same Brand Lock values into every related generation prompt: exact hex, font roles, shape language, placement, clear space, composition, and forbidden treatments.
Use Recraft V4.1 vector mode only for new logo marks.
Use Seedream as the primary photoreal mockup generator. Use GPT Image 2 only for the controlled stage that adds readable text or exact graphic details.
Use local deterministic SVG/PPTX/HTML construction for exact copy and editable layouts. Do not ask an image model to fake editable files.
Do not promise native Figma, Canva, PSD, AI, or EPS files.
Deterministic scripts
Create JSON input files under "$BRANDKIT_WORKDIR"; never interpolate user text directly into shell arguments.
python3 "$SKILL_ROOT/scripts/brandkit.py" preview \
--input "$BRANDKIT_WORKDIR/reviews.json" \
--output-dir "$BRANDKIT_WORKDIR/reviews"
python3 "$SKILL_ROOT/scripts/brandkit.py" logo-export \
--input "$BRANDKIT_WORKDIR/logo-export.json" \
--output-dir "$BRANDKIT_WORKDIR/logo"
python3 "$SKILL_ROOT/scripts/brandkit.py" brandbook-build \
--state-file "$BRANDKIT_STATE" \
--input "$BRANDKIT_WORKDIR/brandbook.json" \
--output-dir "$BRANDKIT_WORKDIR/brandbook"For logo export, load logo export payloads. For Brandbooks, use the bundled builder only; never substitute an improvised PowerPoint or PDF generator after a deterministic contract failure.
Failure policy
Retry a failed Recraft or image-generation request once with the same locked concept and corrected contract. Stop after the second equivalent failure.
If preview or logo export fails twice, report the concrete error; never replace it with ad-hoc SVG rewriting.
If the Brandbook template, font, or conversion contract fails, stop immediately. Do not produce a visually different fallback and call it canonical.
If exact typography or official-logo fidelity cannot be preserved, disclose the limitation instead of claiming completion.
Never expose raw auth tokens or credentials in files, logs, or chat.
Delivery
For Brandbooks, follow the strict response contract in brandbook: PPTX link/path, PDF link/path, and font-install warning only.
For other outputs return:
The requested visual files and previews.
A compact Brand Lock summary.
Editable versus flattened format labels.
Required-font/import limitations.
Stable variant names for targeted revisions.
Reference index
Prerequisites — stage-specific local dependencies and install commands.
Intake — minimal questions and input routing.
Asset analysis — official/reference classification and measurement.
State routing and state payloads — persistent approvals.
Brand Lock — canonical visual constraints.
Design Brain — private art direction.
Concept boards, preview payloads, and inline reviews — selection stages.
Logo, logo prompt enhancer, and logo export payloads — SVG generation and deterministic variants.
Palette and typography — foundation slots.
Mockups, social graphics, posters/banners, packaging, signage, and merchandise — applications.
Presentation decks and Brandbooks — editable documents.
QA and iteration — preflight, repair, approval, and delivery manifest.
Files
- ATTRIBUTION.mdAdded by AutoGineer878 B
- LICENSELicence file added by AutoGineer1.0 KiB
- SKILL.md12.4 KiB
- agents/openai.yaml231 B
- references/asset-analysis.md7.5 KiB
- references/brand-lock.md6.1 KiB
- references/brandbook.md4.8 KiB
- references/brandkit-design-brain.md8.0 KiB
- references/concept-boards.md7.1 KiB
- references/handoff.md3.7 KiB
- references/inline-widgets.md2.7 KiB
- references/intake.md3.3 KiB
- references/logo-export-payloads.md709 B
- references/logo-prompt-enhancer.md9.5 KiB
- references/logo.md10.1 KiB
- references/merchandise.md1.0 KiB
- references/mockups.md10.9 KiB
- references/packaging.md1.3 KiB
- references/palette.md1.0 KiB
- references/posters-banners.md1.1 KiB
- references/prerequisites.md2.9 KiB
- references/presentation-deck.md1.6 KiB
- references/preview-payloads.md2.9 KiB
- references/qa-and-iteration.md6.8 KiB
- references/signage.md1001 B
- references/social-templates.md4.3 KiB
- references/state-payloads.md3.1 KiB
- references/typography.md4.3 KiB
- scripts/brandkit.pyScript54.8 KiB
- scripts/build_brandbook.pyScript36.9 KiB
- scripts/render_brandbook_pdf.pyScript5.2 KiB
references/logo-prompt-enhancer.md
Brandkit Recraft logo prompt enhancer
The final prompt-construction layer between Brandkit Design Brain and Recraft V4.1 vector generation. There is no server-side enhancer tool in this environment: you apply this contract yourself, exactly once per Design Brain candidate, converting one structured logo-candidate specification into one precise Recraft prompt with the highest possible first-pass success rate.
Design Brain has already made the creative decisions. Do not replace, reinterpret, broaden, or add a second concept. Do not narrate this step to the user.
Input contract
Assemble one structured candidate specification per mechanism before writing its prompt:
{
"brand_context": {
"name": "context only; never render it",
"offering": "...",
"industry": "...",
"positioning": "...",
"audience": "...",
"values": ["..."]
},
"visual_axes": {
"restrained_expressive": 50,
"geometric_organic": 50,
"familiar_experimental": 50
},
"candidate": {
"mark_type": "lettermark_monogram | pictorial | abstract | mascot | emblem",
"central_idea": "...",
"visual_mechanism": "...",
"distinctive_element": "specific silhouette, negative-space device, motif treatment, or unexpected locked color pairing",
"shape_logic": "...",
"treatment": "flat_vector | monoline | vector_gradient | hand_drawn_vector",
"style_register": "...",
"user_style_directive": "explicit user-requested style, or null",
"composition": "..."
},
"palette": {
"count": "1, 2, or 3 as required by the locked concept; greater only when user_requested_more_than_three is true",
"user_requested_more_than_three": false,
"roles": ["primary", "accent", "background"]
},
"reference_signals": ["formal qualities only"],
"forbidden_elements": ["..."]
}Treat supplied creative decisions as authoritative. If a nonessential detail is missing, infer the smallest sensible default without changing the central idea.
Output contract
The result of this step is exactly one continuous enhanced Recraft prompt string per candidate — no bullet points inside the prompt, no explanation, no debug text.
The prompt must follow this order:
mark type → central subject/mechanism → shape logic → style register → palette behavior → composition → constraint tail
Every clause must materially affect the drawing.
The central subject/mechanism portion must state exactly one visual idea in one clause. Do not add a second metaphor, alternative, “and/or” construction, or hybrid concept.
Stage boundary — symbol only
This flow creates a symbol/mark before typography selection.
Never include:
brand name
wordmark
tagline
descriptor
invented letters
any other readable words
The only exception is a lettermark/monogram candidate: the explicitly supplied initials inside candidate.visual_mechanism are permitted. For lettermarks and monograms, “no text” means no additional words, taglines, descriptors, or unrelated lettering.
Every candidate constraint tail must include the exact phrase “no text.” It does not need to be the final phrase.
Mark type
Preserve candidate.mark_type exactly.
Allowed types:
lettermark_monogram — explicitly supplied initials or interwoven letterforms
pictorial — one recognizable literal object
abstract — a concept rendered as nonrepresentational geometry
mascot — one character/creature with a clear scalable expression
emblem — a text-free symbol contained within a badge/seal boundary
If mark_type is unexpectedly missing, infer it from visual_mechanism. Default to abstract, never wordmark or combination mark.
Treatment
Preserve candidate.treatment exactly.
Preserve candidate.style_register and candidate.user_style_directive. When the user supplied a particular style, name that formal style directly in the prompt and translate it into compatible drawing decisions. Do not dilute it into a generic “modern,” “minimal,” or “premium” treatment. Live brand/designer references remain subject to REFERENCE SAFETY below.
flat_vector: Solid fills, clean SVG paths, no surface effects.
monoline: Uniform stroke weight, rounded caps, no fills.
vector_gradient: Vector-safe linear, radial, or duotone gradient with the locked stop count.
hand_drawn_vector: Allowed only when supplied explicitly. Preserve intentional stroke variation and a clear silhouette. Do not promise minimal anchor points.
Dimensional/3D treatment is forbidden.
Structural priorities
Define:
one dominant unified silhouette
concrete geometric or organic construction logic
symmetry or intentional asymmetry
positive and negative-space behavior
stroke/fill behavior
internal detail limit
small-size scalability
centered isolated presentation
Prefer one coherent mechanism over several decorative ideas.
Distinctiveness and complexity floor
Every enhanced prompt must contain at least one concrete distinctive element:
a specifically described silhouette
a specific negative-space device
a particular motif treatment
an unexpected but locked color-role pairing
Generic adjectives do not satisfy this requirement. “Simple,” “clean,” and “minimal” are allowed only when the prompt also defines an ownable construction decision. Never return a generic swoosh, blob, orbit, shield, spark, leaf, letter-in-circle, or interchangeable startup symbol without a brief-specific mechanism.
Describe the distinctive element concretely enough that another designer could sketch its structure without guessing.
Preserve candidate.distinctive_element and make it explicit in the prompt.
One concept per logo
The prompt must express one central visual idea only. Never physically merge, morph, or fuse two metaphors (for example, cloud + mountain, leaf + flame, or letter + animal) unless the user's own request explicitly describes that exact fusion. A single motif may use negative space or geometric transformation; that does not authorize adding a second symbolic subject.
Visual axes
Translate axes into drawing decisions:
restrained_expressivecontrols intensity, contrast, and detail densitygeometric_organiccontrols construction, curves, and regularityfamiliar_experimentalcontrols category recognition and novelty
Do not print values or mention axes in the output.
Palette
The palette is locked. Do not invent, replace, expand, or reinterpret it.
Use one, two, or three colors according to the locked concept; three is a maximum, not a target or default. Never add colors merely to reach three. Count the background when it participates visually. More than three are allowed only when palette.user_requested_more_than_three is true. Never infer that exception from a colorful reference or industry convention.
Exact hex values are passed separately through the Recraft request's colors and background_color params. Never put hex, RGB, Pantone, or other color codes in the prompt.
State:
strict color count
role relationships
solid or gradient behavior
background relationship
Use role language such as “locked primary tone,” “locked accent,” and “locked background.” Do not invent color names that were not supplied.
Reference safety
Never output a live brand, studio, artist, or designer name.
Translate reference_signals into formal qualities only: geometry, contrast, density, rhythm, form register, material impression, energy.
Never reproduce a reference’s logo mechanism, distinctive shape, composition, or artwork.
Vector language
For flat_vector, monoline, and vector_gradient, never use: lens, camera, lighting, depth of field, photorealistic, cinematic, material rendering, grain, paper texture, shadows, mockup, scene.
Define drawing logic, not presentation photography.
Constraint tails
flat_vector: Flat vector design, clean lines, no shadows, no texture, no text. Clean editable vector paths, SVG-friendly, minimal anchor points.
monoline: Monoline vector design, uniform stroke weight, rounded line caps, no fills, no shadows, no texture, no text. Clean editable vector paths, SVG-friendly, minimal anchor points.
vector_gradient: Flat vector design with the specified locked vector gradient, no shadows, no texture, no text. Clean editable vector paths, SVG-friendly, minimal anchor points.
hand_drawn_vector: Intentional hand-drawn vector strokes, approved surface variation, clear scalable silhouette, no shadows, no text.
Forbidden elements
Honor every forbidden_elements entry literally. Never replace one forbidden cliché with another generic symbol.
Silent validation
Before submitting each Recraft request, silently verify:
prompt follows the supplied
central_ideaandvisual_mechanismcentral mechanism is one visual idea stated in one clause
no metaphors are fused unless the user explicitly requested that exact fusion
mark_typeandtreatmentare unchangedexplicit user style is present and not generalized away
there is one coherent mechanism
at least one concrete distinctive element clears the complexity floor
no brand/designer names appear
no words appear except explicitly permitted monogram initials
exact phrase “no text” appears in the constraint tail
palette uses at most three colors unless the explicit user override is true
palette is not padded with unnecessary colors
palette count and role behavior are strict
geometry is practical for SVG generation
forbidden elements are absent
no camera or unsupported texture language appears
If any check fails, rewrite the prompt before submitting it.
Requirements and compatibility
Written for: Claude Code
Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta
“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.
Needs: Declares allowed tools, Claude Code frontmatter, Needs network access, Needs Python
Security
What our scanner noticed· Medium risk
- Instructions that use the network
- Downloads piped into a shell
- Scripts that your agent may run
This skill includes scripts that your agent may run. Read them before you use it.
A static scan is not a guarantee. Skills are third-party content: review a skill before you let an agent use it.
Attribution and licence
- Published by
- Higgsfield AI
- Publisher
- Higgsfield AI (Official)
- Source
- higgsfield-ai/skills · higgsfield-brandkit · commit 2700040
- Licence
- MIT License (redistribution allowed)
- LICENSE (added by AutoGineer from LICENSE)
- ATTRIBUTION.md (written by AutoGineer)
- Synced
AutoGineer is not affiliated with or endorsed by Higgsfield AI.
Download ZIP
The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download higgsfield-brandkit.zip