⚠️

重要前提

安装AI Skills的关键前提是：必须科学上网，且开启TUN模式，这一点至关重要，直接决定安装能否顺利完成，在此郑重提醒三遍：科学上网，科学上网，科学上网。查看完整安装教程 →

Bria AI：可控图像生成与编辑工具 - 商业安全AI模型FIBO、RMBG-2.0、GenFill

bria-ai by bria-ai/bria-skill

71 周安装量

51 GitHub Stars

GitHub

安装命令

npx skills add https://github.com/bria-ai/bria-skill --skill bria-ai

AI/机器学习内容创作图像处理

🇨🇳中文介绍

Bria — 可控图像生成与编辑

使用 Bria 的商业安全 AI 模型（FIBO、RMBG-2.0、GenFill 等）生成并精确控制视觉资产。与黑盒生成器不同，Bria 为您提供细粒度控制：通过文本指令编辑、遮罩特定区域、添加/替换/移除单个对象、独立更改光照或季节，并在流水线中链接操作。

设置 — API 密钥检查

在进行任何 Bria API 调用之前，请检查 API 密钥，如果缺失则帮助用户进行设置：

步骤 1：检查密钥是否存在

echo $BRIA_API_KEY

如果输出不为空，请跳至下一节。

步骤 2：如果密钥缺失，引导用户

在浏览器中打开 Bria API 密钥页面：

open "https://platform.bria.ai/console/account/api-keys?utm_source=skill&utm_campaign=bria_skills&utm_content=adjust_photoshop_for_agent"   # macOS
# xdg-open "https://platform.bria.ai/console/account/api-keys?utm_source=skill&utm_campaign=bria_skills&utm_content=adjust_photoshop_for_agent"  # Linux

然后准确告诉用户：

我已在您的浏览器中打开 Bria 网站。要使用图像生成功能，您需要一个免费的 API 密钥。

在我刚刚打开的页面上注册或登录

点击 创建 API 密钥

复制密钥并在此处粘贴

等待用户提供他们的 API 密钥。在用户提供密钥之前，请勿继续。

步骤 3：永久保存密钥

一旦用户提供密钥，请保存它，以便它在不同会话中持久存在。

，检测 shell 并追加到正确的配置文件中：

广告位招租

在这里展示您的产品或服务

触达数万 AI 开发者，精准高效

联系我们

需求	能力	用例
创建新图像	FIBO 生成	主图、产品照片、插图
通过文本编辑	FIBO-编辑	更改颜色、修改对象、变换场景
使用遮罩编辑	GenFill/擦除	精确修复、添加/替换特定区域
添加/替换/移除对象	基于文本的编辑	添加花瓶、用梨替换苹果、移除桌子
透明背景	RMBG-2.0	提取主体用于叠加、徽标、抠图
背景操作	替换/模糊/擦除	更改、模糊或移除背景
扩展图像	外部绘制	扩展边界、更改宽高比
放大图像	超分辨率	提高分辨率 2 倍或 4 倍
增强质量	增强	改善光照、颜色、细节
变换风格	重风格化	油画、动漫、卡通、3D 渲染
更改光照	重光照	黄金时刻、聚光灯、戏剧性光照
更改季节	重季节	春季、夏季、秋季、冬季
混合/合成	图像混合	应用纹理、徽标、合并图像
修复照片	修复	修复老旧/损坏的照片
着色	着色	为黑白照片添加颜色，或转换为黑白
草图转照片	草图转图像	将绘图转换为逼真照片
产品摄影	生活方式拍摄	将产品置于场景中
产品集成	产品集成	将产品嵌入到场景中的精确坐标

移除背景 (RMBG-2.0)

curl -X POST "https://engine.prod.bria-api.com/v2/image/edit/remove_background" \
  -H "api_token: $BRIA_API_KEY" \
  -H "Content-Type: application/json" \
  -H "User-Agent: BriaSkills/1.2.3" \
  -d '{"image": "https://..."}'

返回带有透明度的 PNG。

编辑图像 (FIBO-Edit) - 无需遮罩

curl -X POST "https://engine.prod.bria-api.com/v2/image/edit" \
  -H "api_token: $BRIA_API_KEY" \
  -H "Content-Type: application/json" \
  -H "User-Agent: BriaSkills/1.2.3" \
  -d '{
    "images": ["https://..."],
    "instruction": "change the mug to red"
  }'

使用遮罩编辑图像区域 (GenFill)

curl -X POST "https://engine.prod.bria-api.com/v2/image/edit/gen_fill" \
  -H "api_token: $BRIA_API_KEY" \
  -H "Content-Type: application/json" \
  -H "User-Agent: BriaSkills/1.2.3" \
  -d '{
    "image": "https://...",
    "mask": "https://...",
    "prompt": "what to generate in masked area"
  }'

扩展图像 (外部绘制)

curl -X POST "https://engine.prod.bria-api.com/v2/image/edit/expand" \
  -H "api_token: $BRIA_API_KEY" \
  -H "Content-Type: application/json" \
  -H "User-Agent: BriaSkills/1.2.3" \
  -d '{
    "image": "base64-or-url",
    "aspect_ratio": "16:9",
    "prompt": "coffee shop background, wooden table"
  }'

curl -X POST "https://engine.prod.bria-api.com/v2/image/edit/increase_resolution" \
  -H "api_token: $BRIA_API_KEY" \
  -H "Content-Type: application/json" \
  -H "User-Agent: BriaSkills/1.2.3" \
  -d '{"image": "https://...", "scale": 2}'

产品生活方式拍摄

curl -X POST "https://engine.prod.bria-api.com/v1/product/lifestyle_shot_by_text" \
  -H "api_token: $BRIA_API_KEY" \
  -H "Content-Type: application/json" \
  -H "User-Agent: BriaSkills/1.2.3" \
  -d '{
    "image": "https://product-with-transparent-bg.png",
    "prompt": "modern kitchen countertop, natural morning light"
  }'

将产品集成到场景中

将一个或多个产品放置在场景中的精确坐标处。产品会自动裁剪并与场景的光照和透视相匹配。

curl -X POST "https://engine.prod.bria-api.com/image/edit/product/integrate" \
  -H "api_token: $BRIA_API_KEY" \
  -H "Content-Type: application/json" \
  -H "User-Agent: BriaSkills/1.2.3" \
  -d '{
    "scene": "https://scene-image-url",
    "products": [
      {
        "image": "https://product-image-url",
        "coordinates": {"x": 100, "y": 200, "width": 300, "height": 400}
      }
    ]
  }'

所有端点都返回异步响应：

{
  "request_id": "uuid",
  "status_url": "https://engine.prod.bria-api.com/v2/status/uuid"
}

轮询 status_url 直到 status: "COMPLETED"，然后获取 result.image_url。

import requests, time

def get_result(status_url, api_key):
    headers = {"api_token": api_key, "User-Agent": "BriaSkills/1.2.3"}
    while True:
        r = requests.get(status_url, headers=headers)
        data = r.json()
        if data["status"] == "COMPLETED":
            return data["result"]["image_url"]
        if data["status"] == "FAILED":
            raise Exception(data.get("error"))
        time.sleep(2)

风格："专业产品摄影" 对比 "随意快照"，"扁平设计插图" 对比 "3D 渲染"
光照："柔和自然光"，"影棚灯光"，"戏剧性阴影"
背景："白色影棚"，"渐变"，"模糊的办公室"，"透明"
构图："居中"，"三分法则"，"左侧留白用于文字"
质量关键词："高质量"，"专业"，"商业级"，"4K"，"锐利对焦"
负面提示："模糊，低质量，像素化"，"文字，水印，徽标"

完整端点文档请参见 references/api-endpoints.md。

端点	用途
`POST /v2/image/generate`	从文本生成图像 (FIBO)

通过文本编辑（无需遮罩）

端点	用途
`POST /v2/image/edit`	使用自然语言指令编辑图像
`POST /v2/image/edit/add_object_by_text`	向图像添加对象
`POST /v2/image/edit/replace_object_by_text`	替换图像中的对象
`POST /v2/image/edit/erase_by_text`	按名称移除对象

使用遮罩编辑

端点	用途
`POST /v2/image/edit/gen_fill`	在遮罩区域内修复/生成
`POST /v2/image/edit/erase`	擦除遮罩区域

端点	用途
`POST /v2/image/edit/remove_background`	移除背景 (RMBG-2.0)
`POST /v2/image/edit/replace_background`	替换为 AI 生成的背景
`POST /v2/image/edit/blur_background`	对背景应用模糊
`POST /v2/image/edit/erase_foreground`	移除前景，填充背景
`POST /v2/image/edit/crop_foreground`	围绕主体紧密裁剪

端点	用途
`POST /v2/image/edit/expand`	外部绘制到新的宽高比
`POST /v2/image/edit/enhance`	增强质量和细节
`POST /v2/image/edit/increase_resolution`	放大 2 倍或 4 倍
`POST /v2/image/edit/blend`	混合/合并图像或纹理
`POST /v2/image/edit/reseason`	更改季节/天气
`POST /v2/image/edit/restyle`	变换艺术风格
`POST /v2/image/edit/relight`	修改光照

端点	用途
`POST /v2/image/edit/sketch_to_colored_image`	将草图转换为照片
`POST /v2/image/edit/restore`	修复老旧/损坏的照片
`POST /v2/image/edit/colorize`	为黑白照片着色或转换为黑白

端点	用途
`POST /v1/product/lifestyle_shot_by_text`	通过文本将产品置于场景中
`POST /image/edit/product/integrate`	将产品集成到场景中的精确坐标

端点	用途
`POST /v2/structured_instruction/generate`	生成 JSON 指令（无图像）
`GET /v2/status/{id}`	检查异步请求状态

所有请求都需要以下请求头：

api_token: YOUR_BRIA_API_KEY
User-Agent: BriaSkills/<version>

必需： 在每个 Bria API 调用中始终包含 User-Agent: BriaSkills/<version> 请求头（其中 <version> 是当前技能版本，例如 BriaSkills/1.2.3）。这适用于所有请求，包括状态轮询。

API 端点参考 — 包含请求/响应格式的完整端点文档
工作流与流水线 — 批量生成、并行流水线、集成示例
Python 客户端 — 功能齐全的异步 Python 客户端
TypeScript 客户端 — 类型化的 Node.js 客户端
Bash/curl 参考 — 所有端点的 Shell 函数

vgl — 编写结构化 VGL JSON 提示，以精确、确定性地控制 FIBO 图像生成
image-utils — 用于后处理的经典图像处理（调整大小、裁剪、合成、水印）

🇺🇸English

Bria — Controllable Image Generation & Editing

Generate and precisely control visual assets using Bria's commercially-safe AI models (FIBO, RMBG-2.0, GenFill, and more). Unlike black-box generators, Bria gives you fine-grained control: edit by text instruction, mask specific regions, add/replace/remove individual objects, change lighting or season independently, and chain operations in pipelines.

Setup — API Key Check

Before making any Bria API call, check for the API key and help the user set it up if missing:

Step 1: Check if the key exists

echo $BRIA_API_KEY

If the output is not empty , skip to the next section.

Step 2: If the key is missing, guide the user

Open the Bria API keys page in the browser:

open "https://platform.bria.ai/console/account/api-keys?utm_source=skill&utm_campaign=bria_skills&utm_content=adjust_photoshop_for_agent"   # macOS
# xdg-open "https://platform.bria.ai/console/account/api-keys?utm_source=skill&utm_campaign=bria_skills&utm_content=adjust_photoshop_for_agent"  # Linux

Then tell the user exactly this:

I opened the Bria website in your browser. To use image generation, you need a free API key.

Sign up or log in on the page I just opened

Click Create API Key

Copy the key and paste it here

Wait for the user to provide their API key. Do not proceed until they give you the key.

Step 3: Save the key permanently

Once the user provides the key, save it so it persists across sessions.

On macOS/Linux , detect the shell and append to the correct profile:

# Detect the shell profile
if [ -n "$ZSH_VERSION" ] || [ "$SHELL" = */zsh ]; then
  PROFILE_FILE="$HOME/.zshrc"
elif [ -f "$HOME/.bashrc" ]; then
  PROFILE_FILE="$HOME/.bashrc"
else
  PROFILE_FILE="$HOME/.profile"
fi

# Append the export (only if not already present)
grep -q 'export BRIA_API_KEY=' "$PROFILE_FILE" 2>/dev/null && \
  sed -i.bak '/export BRIA_API_KEY=/d' "$PROFILE_FILE"
echo 'export BRIA_API_KEY="THE_KEY_THE_USER_GAVE_YOU"' >> "$PROFILE_FILE"

# Make it available immediately in this session
export BRIA_API_KEY="THE_KEY_THE_USER_GAVE_YOU"

On Windows (PowerShell) :

[System.Environment]::SetEnvironmentVariable("BRIA_API_KEY", "THE_KEY_THE_USER_GAVE_YOU", "User")
$env:BRIA_API_KEY = "THE_KEY_THE_USER_GAVE_YOU"

Replace THE_KEY_THE_USER_GAVE_YOU with the actual key the user provided.

Step 4: Verify

echo $BRIA_API_KEY

Confirm the key is set, then tell the user:

Your API key is saved and will persist across sessions. You're all set!

Do not proceed with any image generation or editing until the API key is confirmed set.

Core Capabilities

Need	Capability	Use Case
Create new images	FIBO Generate	Hero images, product shots, illustrations
Edit by text	FIBO-Edit	Change colors, modify objects, transform scenes
Edit with mask	GenFill/Erase	Precise inpainting, add/replace specific regions
Add/Replace/Remove objects	Text-based editing	Add vase, replace apple with pear, remove table
Transparent backgrounds	RMBG-2.0	Extract subjects for overlays, logos, cutouts
Background operations	Replace/Blur/Erase	Change, blur, or remove backgrounds
Expand images	Outpainting	Extend boundaries, change aspect ratios
Upscale images	Super Resolution	Increase resolution 2x or 4x
Enhance quality

Quick Reference

Generate an Image (FIBO)

curl -X POST "https://engine.prod.bria-api.com/v2/image/generate" \
  -H "api_token: $BRIA_API_KEY" \
  -H "Content-Type: application/json" \
  -H "User-Agent: BriaSkills/1.2.3" \
  -d '{
    "prompt": "your description",
    "aspect_ratio": "16:9",
    "resolution": "1MP"
  }'

Aspect ratios : 1:1 (square), 16:9 (hero/banner), 4:3 (presentation), 9:16 (mobile/story), 3:4 (portrait)

Resolution : 1MP (default) or 4MP (improved details for photorealism, adds ~30s latency)

Advanced : For precise, deterministic control over generation, use VGL structured prompts instead of natural language. VGL defines every visual attribute (objects, lighting, composition) as explicit JSON.

Remove Background (RMBG-2.0)

curl -X POST "https://engine.prod.bria-api.com/v2/image/edit/remove_background" \
  -H "api_token: $BRIA_API_KEY" \
  -H "Content-Type: application/json" \
  -H "User-Agent: BriaSkills/1.2.3" \
  -d '{"image": "https://..."}'

Returns PNG with transparency.

Edit Image (FIBO-Edit) - No Mask Required

curl -X POST "https://engine.prod.bria-api.com/v2/image/edit" \
  -H "api_token: $BRIA_API_KEY" \
  -H "Content-Type: application/json" \
  -H "User-Agent: BriaSkills/1.2.3" \
  -d '{
    "images": ["https://..."],
    "instruction": "change the mug to red"
  }'

Edit Image Region with Mask (GenFill)

curl -X POST "https://engine.prod.bria-api.com/v2/image/edit/gen_fill" \
  -H "api_token: $BRIA_API_KEY" \
  -H "Content-Type: application/json" \
  -H "User-Agent: BriaSkills/1.2.3" \
  -d '{
    "image": "https://...",
    "mask": "https://...",
    "prompt": "what to generate in masked area"
  }'

Expand Image (Outpainting)

curl -X POST "https://engine.prod.bria-api.com/v2/image/edit/expand" \
  -H "api_token: $BRIA_API_KEY" \
  -H "Content-Type: application/json" \
  -H "User-Agent: BriaSkills/1.2.3" \
  -d '{
    "image": "base64-or-url",
    "aspect_ratio": "16:9",
    "prompt": "coffee shop background, wooden table"
  }'

Upscale Image

curl -X POST "https://engine.prod.bria-api.com/v2/image/edit/increase_resolution" \
  -H "api_token: $BRIA_API_KEY" \
  -H "Content-Type: application/json" \
  -H "User-Agent: BriaSkills/1.2.3" \
  -d '{"image": "https://...", "scale": 2}'

Product Lifestyle Shot

curl -X POST "https://engine.prod.bria-api.com/v1/product/lifestyle_shot_by_text" \
  -H "api_token: $BRIA_API_KEY" \
  -H "Content-Type: application/json" \
  -H "User-Agent: BriaSkills/1.2.3" \
  -d '{
    "image": "https://product-with-transparent-bg.png",
    "prompt": "modern kitchen countertop, natural morning light"
  }'

Integrate Products into Scene

Place one or more products at exact coordinates in a scene. Products are automatically cut out and matched to the scene's lighting and perspective.

curl -X POST "https://engine.prod.bria-api.com/image/edit/product/integrate" \
  -H "api_token: $BRIA_API_KEY" \
  -H "Content-Type: application/json" \
  -H "User-Agent: BriaSkills/1.2.3" \
  -d '{
    "scene": "https://scene-image-url",
    "products": [
      {
        "image": "https://product-image-url",
        "coordinates": {"x": 100, "y": 200, "width": 300, "height": 400}
      }
    ]
  }'

Async Response Handling

All endpoints return async responses:

{
  "request_id": "uuid",
  "status_url": "https://engine.prod.bria-api.com/v2/status/uuid"
}

Poll the status_url until status: "COMPLETED", then get result.image_url.

import requests, time

def get_result(status_url, api_key):
    headers = {"api_token": api_key, "User-Agent": "BriaSkills/1.2.3"}
    while True:
        r = requests.get(status_url, headers=headers)
        data = r.json()
        if data["status"] == "COMPLETED":
            return data["result"]["image_url"]
        if data["status"] == "FAILED":
            raise Exception(data.get("error"))
        time.sleep(2)

Prompt Engineering Tips

Style : "professional product photography" vs "casual snapshot", "flat design illustration" vs "3D rendered"
Lighting : "soft natural light", "studio lighting", "dramatic shadows"
Background : "white studio", "gradient", "blurred office", "transparent"
Composition : "centered", "rule of thirds", "negative space on left for text"
Quality keywords : "high quality", "professional", "commercial grade", "4K", "sharp focus"
Negative prompts : "blurry, low quality, pixelated", "text, watermark, logo"

API Reference

See references/api-endpoints.md for complete endpoint documentation.

Key Endpoints

Generation

Endpoint	Purpose
`POST /v2/image/generate`	Generate images from text (FIBO)

Edit by Text (No Mask)

Endpoint	Purpose
`POST /v2/image/edit`	Edit image with natural language instruction
`POST /v2/image/edit/add_object_by_text`	Add objects to image
`POST /v2/image/edit/replace_object_by_text`	Replace objects in image
`POST /v2/image/edit/erase_by_text`	Remove objects by name

Edit with Mask

Endpoint	Purpose
`POST /v2/image/edit/gen_fill`	Inpaint/generate in masked region
`POST /v2/image/edit/erase`	Erase masked region

Background Operations

Endpoint	Purpose
`POST /v2/image/edit/remove_background`	Remove background (RMBG-2.0)
`POST /v2/image/edit/replace_background`	Replace with AI-generated background
`POST /v2/image/edit/blur_background`	Apply blur to background
`POST /v2/image/edit/erase_foreground`	Remove foreground, fill background
`POST /v2/image/edit/crop_foreground`	Crop tightly around subject

Image Transformation

Endpoint	Purpose
`POST /v2/image/edit/expand`	Outpaint to new aspect ratio
`POST /v2/image/edit/enhance`	Enhance quality and details
`POST /v2/image/edit/increase_resolution`	Upscale 2x or 4x
`POST /v2/image/edit/blend`	Blend/merge images or textures
`POST /v2/image/edit/reseason`	Change season/weather
`POST /v2/image/edit/restyle`

Restoration

Endpoint	Purpose
`POST /v2/image/edit/sketch_to_colored_image`	Convert sketch to photo
`POST /v2/image/edit/restore`	Restore old/damaged photos
`POST /v2/image/edit/colorize`	Colorize B&W or convert to B&W

Product Photography

Endpoint	Purpose
`POST /v1/product/lifestyle_shot_by_text`	Place product in scene by text
`POST /image/edit/product/integrate`	Integrate products into scene at exact coordinates

Utilities

Endpoint	Purpose
`POST /v2/structured_instruction/generate`	Generate JSON instruction (no image)
`GET /v2/status/{id}`	Check async request status

Authentication

All requests need these headers:

api_token: YOUR_BRIA_API_KEY
User-Agent: BriaSkills/<version>

Required: Always include the User-Agent: BriaSkills/<version> header in every Bria API call (where <version> is the current skill version, e.g. BriaSkills/1.2.3). This applies to all requests including status polling.

Additional Resources

API Endpoints Reference — Complete endpoint documentation with request/response formats
Workflows& Pipelines — Batch generation, parallel pipelines, integration examples
Python Client — Full-featured async Python client
TypeScript Client — Typed Node.js client
Bash/curl Reference — Shell functions for all endpoints

Related Skills

vgl — Write structured VGL JSON prompts for precise, deterministic control over FIBO image generation
image-utils — Classic image manipulation (resize, crop, composite, watermarks) for post-processing

Weekly Installs

Repository

bria-ai/bria-skill

GitHub Stars

First Seen

Feb 8, 2026

Security Audits

Gen Agent Trust HubPass SocketPass SnykFail

Installed on

codex32

opencode31

gemini-cli31

github-copilot31

amp31

kimi-cli31

AI界面设计评审工具 - 全面评估UI/UX设计质量、检测AI生成痕迹与优化用户体验

58,500 周安装