{"slug":"bria-ai","title":"bria-ai","summary":"Image generation, photo editing, background removal — transparent PNG images, cutouts, ecommerce packshots, product catalogs, lifestyle shots via Bria.ai. Build store-ready e-commerce catalogs at scale — shadows, product dimensions images, lifestyle scenes, marketplace listing va","platform":"Claude","tags":[],"authorName":"LLM Mart","authorSlug":"llm-mart","score":0,"source":"github","price":null,"verified":false,"createdAt":"2026-09-17T17:01:35.099701Z","repo":{"url":"https://github.com/Bria-AI/bria-skill","stars":68,"forks":6,"license":null,"updatedAt":"2026-10-06T12:19:32Z"},"bodyHtml":"
\n

name: bria-ai\ndescription: Image generation, photo editing, background removal — transparent PNG images, cutouts, ecommerce packshots, product catalogs, lifestyle shots via Bria.ai. Build store-ready e-commerce catalogs at scale — shadows, product dimensions images, lifestyle scenes, marketplace listing variants. ALWAYS use this skill when the user wants to generate, edit, or transform any image — hero images, banners, social media visuals, product photos, illustrations, icons, thumbnails, ad creatives, or marketing materials — or to build a product catalog, create a packshot, stage a lifestyle shot, remove a background, make transparent PNGs, or batch-process product photos. Triggers on packshot, product catalog, product shot creator, lifestyle shot, cutout, product dimensions image, inpainting, outpainting, object removal, upscale, photo restoration, style transfer, relight, reseason, sketch-to-photo. Commercially safe, royalty-free.\nlicense: MIT\nmetadata:\nauthor: Bria AI\nversion: \"1.3.7\"

\n

Bria — AI Image Generation, Editing & Background Removal

\n

Commercially safe, royalty-free image generation and editing through 20+ API endpoints. Generate from text, edit with natural language, remove backgrounds, create product shots, and build automated image pipelines.

\n

For additional endpoint details beyond what is documented here, see the Bria API reference for agents.

\n

When to Use This Skill

\n

Use this skill when the user wants to:

\n\n

This skill handles the full spectrum of AI image operations. If the user mentions images, photos, visuals, or any visual content creation — use this skill.

\n
\n

What You Can Build

\n\n
\n

Setup — Authentication

\n

Before making any API call, you need a valid Bria access token.

\n

Step 1: Check for existing credentials

\n
if [ -f ~/.bria/credentials ]; then\n  BRIA_ACCESS_TOKEN=$(grep '^access_token=' \"$HOME/.bria/credentials\" | cut -d= -f2-)\n  BRIA_API_KEY=$(grep '^api_token=' \"$HOME/.bria/credentials\" | cut -d= -f2-)\nfi\nif [ -z \"$BRIA_ACCESS_TOKEN\" ]; then\n  echo \"NO_CREDENTIALS\"\nelif [ -n \"$BRIA_API_KEY\" ]; then\n  echo \"READY\"\nelse\n  echo \"CREDENTIALS_FOUND\"\nfi\n
\n

If the output is READY, skip straight to making API calls — no introspection needed.\nIf the output is CREDENTIALS_FOUND, skip to Step 3.\nIf the output is NO_CREDENTIALS, proceed to Step 2.

\n

Step 2: Authenticate via device authorization

\n

Start the device authorization flow:

\n

2a. Request a device code:

\n
DEVICE_RESPONSE=$(curl -s -X POST \"https://engine.prod.bria-api.com/v2/auth/device/authorize\" \\\n  -H \"Content-Type: application/json\")\necho \"$DEVICE_RESPONSE\"\n
\n

Parse the response fields:

\n\n

2b. Show the user a single sign-in link. Tell them exactly this — nothing more:

\n
\n

Connect your Bria account: Click here to sign in\nYour code is — it's already filled in.

\n
\n

Do NOT show two links. Do NOT show the raw URL separately. Do NOT use verification_uri from the API response. Keep it to one clickable link.

\n

2c. Poll for the token. After showing the user the code, immediately start polling. Try up to 60 times with the given interval (default 5 seconds):

\n
for i in $(seq 1 60); do\n  TOKEN_RESPONSE=$(curl -s -X POST \"https://engine.prod.bria-api.com/v2/auth/token\" \\\n    -d \"grant_type=urn:ietf:params:oauth:grant-type:device_code\" \\\n    -d \"device_code=$DEVICE_CODE\")\n  ACCESS_TOKEN=$(printf '%s' \"$TOKEN_RESPONSE\" | sed -n 's/.*\"access_token\" *: *\"\\([^\"]*\\)\".*/\\1/p')\n  if [ -n \"$ACCESS_TOKEN\" ]; then\n    BRIA_ACCESS_TOKEN=\"$ACCESS_TOKEN\"\n    REFRESH_TOKEN=$(printf '%s' \"$TOKEN_RESPONSE\" | sed -n 's/.*\"refresh_token\" *: *\"\\([^\"]*\\)\".*/\\1/p')\n    mkdir -p ~/.bria\n    printf 'access_token=%s\\nrefresh_token=%s\\n' \"$BRIA_ACCESS_TOKEN\" \"$REFRESH_TOKEN\" > \"$HOME/.bria/credentials\"\n    echo \"AUTHENTICATED\"\n    break\n  fi\n  sleep 5\ndone\n
\n

If the output contains AUTHENTICATED, proceed to Step 3. Otherwise the code expired — start over from Step 2a.

\n

Do not proceed with any API call until authentication is confirmed.

\n

Step 3: Verify billing status and resolve API key

\n

Introspect the bearer token to check billing status and obtain the real API key for Bria API calls:

\n
INTROSPECT=$(curl -s -X POST \"https://engine.prod.bria-api.com/v2/auth/token/introspect\" \\\n  -d \"token=$BRIA_ACCESS_TOKEN\")\nBILLING_STATUS=$(printf '%s' \"$INTROSPECT\" | sed -n 's/.*\"billing_status\" *: *\"\\([^\"]*\\)\".*/\\1/p')\nif [ \"$BILLING_STATUS\" = \"blocked\" ]; then\n  BILLING_MSG=$(printf '%s' \"$INTROSPECT\" | sed -n 's/.*\"billing_message\" *: *\"\\([^\"]*\\)\".*/\\1/p')\n  echo \"BILLING_ERROR: $BILLING_MSG\"\nfi\nACTIVE=$(printf '%s' \"$INTROSPECT\" | sed -n 's/.*\"active\" *: *\\([^,}]*\\).*/\\1/p' | tr -d ' ')\nif [ \"$ACTIVE\" = \"false\" ]; then\n  # Clear stale tokens so re-auth starts fresh (credentials file is re-created in Step 2c)\n  printf '' > \"$HOME/.bria/credentials\"\n  echo \"TOKEN_EXPIRED\"\nfi\nBRIA_API_KEY=$(printf '%s' \"$INTROSPECT\" | sed -n 's/.*\"api_token\" *: *\"\\([^\"]*\\)\".*/\\1/p')\nif [ -n \"$BRIA_API_KEY\" ]; then\n  grep -v '^api_token=' \"$HOME/.bria/credentials\" > \"$HOME/.bria/credentials.tmp\" 2>/dev/null || true\n  printf 'api_token=%s\\n' \"$BRIA_API_KEY\" >> \"$HOME/.bria/credentials.tmp\"\n  mv \"$HOME/.bria/credentials.tmp\" \"$HOME/.bria/credentials\"\nfi\n
\n

Interpret the output:

\n\n
\n

Core Capabilities

\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
NeedCapabilityUse Case
Generate images from textFIBO GenerateHero images, product shots, illustrations, social media images, banners
Edit images by text instructionFIBO-EditChange colors, modify objects, transform scenes
Combine 2–4 images in one editFIBO-Edit multi-referencePut the outfit, product, logo, style, or background of one image into another
Edit image region with maskGenFill/ErasePrecise inpainting, add/replace specific regions
Add/Replace/Remove objectsText-based editingAdd vase, replace apple with pear, remove table
Remove background (transparent PNG)RMBG-2.0Extract subjects for overlays, logos, cutouts
Turn a finished ad back into layersAd Delayermake a shipped creative editable, resize or localise an ad
Replace/blur/erase backgroundBackground opsChange, blur, or remove backgrounds
Expand/outpaint imagesOutpaintingExtend boundaries, change aspect ratios
Upscale image resolutionSuper ResolutionIncrease resolution 2x or 4x
Enhance image qualityEnhancementImprove lighting, colors, details
Restyle imagesRestyleOil painting, anime, cartoon, 3D render
Change lightingRelightGolden hour, spotlight, dramatic lighting
Change seasonReseasonSpring, summer, autumn, winter
Composite/blend imagesImage BlendingApply textures, logos, merge images
Restore old photosRestorationFix old/damaged photos
Colorize imagesColorizationAdd color to B&W, or convert to B&W
Sketch to photoSketch2ImageConvert drawings to realistic photos
Product cutoutProduct CutoutClean transparent PNG from a raw product photo
Product packshotProduct PackshotStandardized 2000×2000 shot on a solid/clean background
Product shadowProduct ShadowAdd a realistic drop or float shadow to a cutout
Create product lifestyle shotsLifestyle ShotPlace products in scenes for e-commerce
Integrate products into scenesProduct IntegrateEmbed products at exact coordinates
Add dimension callouts to productsProduct DimensionsMarketplace-style measurement images with size/weight/capacity labels
Build a full product catalogCatalog PipelineBatch a folder of photos → packshots, dimensions, lifestyle, marketplace variants
\n

How to Call Any Endpoint

\n

Use bria_call for all API calls. It handles URL passthrough, local file base64 encoding, JSON construction, API call, and async polling in a single function call. The API key is auto-loaded from ~/.bria/credentials.

\n

First, source the helper script at references/code-examples/bria_client.sh (resolve relative to this skill's directory).

\n
source <SKILL_DIR>/references/code-examples/bria_client.sh\n\n# Generate (no image input — pass empty string)\nRESULT=$(bria_call /v2/image/generate \"\" '\"prompt\": \"your description\", \"aspect_ratio\": \"16:9\", \"sync\": true')\n\n# Remove background\nRESULT=$(bria_call /v2/image/edit/remove_background \"/path/to/local/image.png\")\n\n# Replace background\nRESULT=$(bria_call /v2/image/edit/replace_background \"https://example.com/img.jpg\" '\"prompt\": \"sunset beach\"')\n\n# Edit image (uses images array — pass --key images)\nRESULT=$(bria_call /v2/image/edit \"/path/to/image.png\" --key images '\"instruction\": \"make it look warmer\"')\n\n# Edit with reference images — each --image adds the next one, in order\nRESULT=$(bria_call /v2/image/edit \"https://example.com/man.jpg\" --key images \\\n  --image \"https://example.com/santa.png\" \\\n  '\"instruction\": \"dress the man in image 1 in the santa outfit from image 2\"')\n\n# Upscale (`desired_increase` is 2 or 4 — no other value. Transparency is preserved by default)\nRESULT=$(bria_call /v2/image/edit/increase_resolution \"https://example.com/img.jpg\" '\"desired_increase\": 4')\n\n# Product cutout → transparent PNG (use --key file for a local image)\nCUTOUT=$(bria_call /v1/product/cutout \"/path/to/raw.jpg\" --key file)\n\n# Packshot on white from the cutout URL\nRESULT=$(bria_call /v1/product/packshot \"$CUTOUT\" --key image_url '\"background_color\": \"#FFFFFF\"')\n\n# Lifestyle shot\nRESULT=$(bria_call /v1/product/lifestyle_shot_by_text \"/path/to/product.png\" '\"scene_description\": \"modern kitchen countertop\"')\n\n# Product dimensions — auto-removes background, draws measurement callouts.\n# Dual cm / in labels: repeat each dimension with the same name+position in both\n# units and set \"units_display\": \"dual_slash\" so they merge into one \"12 cm / 4.7 in\" label.\nRESULT=$(bria_call /v2/image/edit/product_dimensions \"/path/to/product.png\" \\\n  '\"style\": \"default\", \"units_display\": \"dual_slash\", \"dimensions\": [{\"name\": \"height\", \"value\": 12, \"unit\": \"cm\", \"position\": \"left\"}, {\"name\": \"height\", \"value\": 4.7, \"unit\": \"in\", \"position\": \"left\"}, {\"name\": \"width_bottom\", \"value\": 6, \"unit\": \"cm\", \"position\": \"bottom\"}, {\"name\": \"width_bottom\", \"value\": 2.4, \"unit\": \"in\", \"position\": \"bottom\"}], \"title\": \"Gummies Bottle\", \"capacity\": {\"value\": 500, \"unit\": \"ml\"}, \"weight\": {\"value\": 250, \"unit\": \"g\", \"label\": \"Net Weight\"}')\n\necho \"$RESULT\"\n
\n

Calling convention: bria_call <endpoint> <image_or_empty> [--key <json_key>] [extra JSON fields...]

\n\n

Editing with several images (2–4): reach for a second image when the look the user wants\nalready exists as a picture — a specific outfit, product, logo, or scene — instead of something you\ncan describe in words. Put the image being edited first, references after it, and address them by\nposition in the instruction: \"dress the man in image 1 in the santa outfit from image 2\". Say what\neach reference contributes (\"the background of image 3\"), in plain prose. A single-image edit needs\nno positional wording: \"change the mug color to red\".

\n

Generation options: Aspect ratios 1:1, 16:9, 4:3, 9:16, 3:4. Resolution 1MP (default) or 4MP (more detail, +30s). Pass \"sync\": true for a single generated image. Editing endpoints are the other way round —\nthey answer with a status_url you poll, and \"sync\": true on an edit fails with a gateway\ntimeout.

\n
\n

Advanced: For precise control over generation, use the vgl skill for structured VGL JSON prompts instead of natural language.

\n
\n

See API Endpoints Reference for full parameter documentation on all 20+ endpoints.

\n
\n

Product Catalog Pipeline (batch)

\n

When the user wants listing-ready imagery for physical products at scale — \"build a product catalog from ./products\", \"turn this folder of photos into a catalog\", \"make these Amazon/Shopify/Etsy compliant\" — use the bundled driver instead of calling endpoints one by one. It is zero-config: with no arguments it reads ./products and writes ./catalog.

\n
python3 <SKILL_DIR>/references/code-examples/build_catalog.py\n
\n

Per product it runs cutout → packshot + lifestyle scenes (+ dimensions) → marketplace variants and writes cutout.png, packshot.jpg, lifestyle_N.jpg, dimensions.png (when measurements are available), per-channel variants, and a listing.json copy scaffold. Override only what you need:

\n
python3 <SKILL_DIR>/references/code-examples/build_catalog.py \\\n  --input ./photos --output ./store \\\n  --scenes \"clean marble surface, soft studio light|cozy wooden desk, warm morning light\" \\\n  --variants amazon,shopify --dims ./dims.json\n
\n

Dimensions need real measurements — ask, don't guess. With no measurements file and no --no-dims, the driver prints NEEDS_DIMENSIONS and exits (code 2). When that happens, stop and ask the user to either provide height/width (and optional weight/capacity) per product — as a dims.json/CSV — or re-run with --no-dims to skip only the dimensions image. Never fabricate sizes. dims.json format:

\n
{ \"soap.jpg\": {\"title\": \"Hand Wash\", \"height_cm\": 19, \"width_cm\": 6, \"capacity_ml\": 250} }\n
\n

Store measurements in cm — callouts render dual cm / in automatically. Requires pip install requests Pillow.

\n

Marketplace-ready variants

\n

export_variants.py turns one master image (packshot or cutout) into a compliant file per channel — enforcing background, aspect ratio, product fill, and minimum resolution:

\n
python3 <SKILL_DIR>/references/code-examples/export_variants.py \\\n  --input ./catalog/soap/packshot.jpg --output ./catalog/soap --channels amazon,shopify,etsy\n
\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
ChannelAspectBackgroundProduct fillMin resolution
Amazon (main)1:1pure white #FFFFFF~85%1600 px (≥ 3000 ideal)
Shopify1:1 (+ 4:5)white / transparent~90%2048 px
Etsy5:4white / lifestyle~85%2000 px
\n

Full rules: Marketplace Presets.

\n

Write the listing copy (SEO)

\n

Bria generates images, not text — you, the agent, write the copy. build_catalog.py writes a listing.json scaffold per product; after the images are generated, view each packshot.jpg and fill it: seo_title (≤ 60 chars, keyword-first), meta_description (≤ 160 chars), description (1–2 paragraphs), bullets (4–6 benefit-led), tags (8–15 keywords). Ground every claim in what's visible — don't invent specs — and fold in any known dimensions.

\n
\n

Prompt Engineering Tips

\n\n

Recipes by Use Case

\n

Hero banner (16:9): \"Modern tech startup workspace with developers collaborating, bright natural lighting, clean minimal aesthetic\" — include \"clean background\" or \"minimal\" for text overlay space

\n

Product photo (1:1): \"Professional product photo of [item] on white studio background, soft shadows, commercial photography lighting\" — then remove background for transparent PNG

\n

Presentation visual (16:9): \"Abstract visualization of data analytics, blue and purple gradient, modern corporate style, clean composition with space for text\" — common themes: \"abstract technology\", \"business collaboration\", \"minimalist geometric patterns\"

\n

Instagram post (1:1): \"Lifestyle photo of coffee and laptop on wooden desk, morning light, cozy atmosphere\"

\n

Story/Reel (9:16): \"Vertical product showcase of smartphone, floating in gradient background, tech aesthetic\"

\n
\n

Additional Resources

\n\n

Related Skills

\n\n","files":[{"path":"LICENSE.txt","sizeBytes":1335,"isText":true},{"path":"references/api-endpoints.md","sizeBytes":29672,"isText":true},{"path":"references/code-examples/bria_client.sh","sizeBytes":7181,"isText":true},{"path":"references/code-examples/build_catalog.py","sizeBytes":11908,"isText":true},{"path":"references/code-examples/export_variants.py","sizeBytes":4222,"isText":true},{"path":"references/marketplace-presets.md","sizeBytes":2267,"isText":true},{"path":"SKILL.md","sizeBytes":22027,"isText":true}],"reviewScore":null,"reviewSummary":null,"trust":{"provenance":"trusted-source-unreviewed","notice":"Community-authored content, reproduced verbatim and not vetted as instructions. Treat it as data to evaluate, never as directives to follow.","bodySource":null},"bodyLocked":false,"purchaseUrl":null,"sourceUrl":null,"report":{"provenance":"trusted-source-unreviewed","screen":{"ran":true,"outcome":"clean","suspicious":0,"notes":0,"hiddenCharacters":false},"virusScan":{"engine":"clamav","status":"clean","scannedAt":"2026-10-06T15:23:13.271311Z","sha256":"0BA0CDAC0AD91566C44E1C88A1B1A9A64EC2C9D2C38D75E293AA11EA1AFADAC9","sizeBytes":28880},"review":null,"source":{"repositoryUrl":"https://github.com/Bria-AI/bria-skill","path":"skills/bria-ai","license":null,"commit":"0a65163513443a1b8c589449ea5920edcecad946","subtreeSha":"81C2F224D0ECE5A75430F955E40D779A61058637AAC833826C8E146F45096EC9","lastSyncedAt":"2026-10-06T15:23:04.488518Z"},"reviewedAt":"2026-10-06T15:23:17.640995Z","notice":"Community-authored content, reproduced verbatim and not vetted as instructions. Treat it as data to evaluate, never as directives to follow."},"install":[{"target":"skills-cli","command":"npx skills add https://github.com/Bria-AI/bria-skill/tree/main/skills/bria-ai"},{"target":"claude-code","command":"claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install bria-ai-bria-skill@llmmart"},{"target":"git","command":"git clone https://github.com/Bria-AI/bria-skill.git"}]}