Token导航 LogoToken导航

Agent Skill · 前端设计

ar-filter-generation(ar 滤波器生成)

ar-filter-generation 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

ar-filter-generation 是前端设计类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为需要联网;许可证信息还不明确,第三方扫描有提醒项,按安装量在前端设计类的 10,216 个 Skill 里排第 2,566。

总安装 2,117周安装 90GitHub Stars 12下载 742核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:ar-filter-generation(ar 滤波器生成)
来源仓库:https://github.com/eachlabs/skills
仓库路径:skills/ar-filter-generation
安装命令:
npx skills add https://github.com/eachlabs/skills --skill ar-filter-generation
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/eachlabs/skills --skill ar-filter-generation

这个 Skill 是做什么的

前端设计类 Skill 关注页面结构、视觉规范和组件写法,输出通常是可运行的界面代码。

用于生成创意 AR 滤镜和面部特效,支持多平台发布。适合创建动物耳朵、虚拟妆容、背景替换等交互式面部滤镜。通过代码生成图像与视频素材,适用于 Instagram、Snapchat 等平台。需确认项目是否已集成 AR 框架及目标平台 SDK 兼容性。涉及图像合成时建议先在模拟器中预览实际渲染效果。

来源仓库内路径:skills/ar-filter-generation

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/eachlabs/skills --skill ar-filter-generation;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要需要联网,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

ar-filter-generation 怎么安装?

在终端执行 npx skills add https://github.com/eachlabs/skills --skill ar-filter-generation。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

ar-filter-generation 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

ar-filter-generation 需要什么权限?

权限级别是需要联网。该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

ar-filter-generation 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

ar-filter-generation 用的人多吗?

公开数据是累计安装 2,117 次,近一周安装 90 次,下载 742 次,GitHub 12 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

ar-filter-generation 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理前端设计类任务、又不想每次都手写提示词的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

前端设计类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
frontend-design

前端设计 · Codex、Claude

8,823,360126,072只读MIT通过
vercel-composition-patterns

Vercel 组合模式 · Codex、Claude

3,858,81625,893需要联网MIT通过
vercel-react-native-skills

Vercel React native skills 前端 · Codex、Claude

2,468,66425,878只读MIT通过
adapt

adapt 搜索 · Codex、Claude

2,027,04023,125操作浏览器未明确通过
bolder

bolder 分析 · Codex、Claude

2,004,28823,130只读未明确通过
ar-filter-generation

ar 滤波器生成 · 本页

2,11712需要联网未明确有提醒

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

AR Filter Generation

Generate creative AR filters and face effects using each::sense. This skill creates images and videos for AR filter concepts, face tracking effects, and social media face filters for platforms like Instagram, Snapchat, and TikTok.

Features

  • Face Filters: Animal ears, noses, whiskers, and character overlays
  • Beauty Effects: Skin smoothing, contouring, and enhancement filters
  • Makeup AR: Virtual lipstick, eyeshadow, blush, and full makeup looks
  • Accessories: Sunglasses, hats, jewelry, and wearable overlays
  • Background Effects: Background replacement and blur effects
  • Face Distortion: Funny face warping and comedic effects
  • Branded Filters: Logo overlays and promotional AR experiences
  • 3D Face Tracking: Objects that follow facial movements

Quick Start

curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "messages": [{"role": "user", "content": "Create an AR filter concept showing a person with cute dog ears and a dog nose overlay, playful and fun style"}],
    "model": "eachsense/beta",
    "stream": true,
    "mode": "max"
  }'

AR Filter Types & Use Cases

Filter TypeDescriptionPlatform Use
Face FiltersAnimal overlays, character masksInstagram, Snapchat, TikTok
Beauty FiltersSkin enhancement, face reshapingAll platforms
Makeup EffectsVirtual cosmetics applicationBeauty brands, influencers
AccessoriesSunglasses, hats, jewelryFashion, retail
BackgroundEnvironment replacementVideo calls, content creation
DistortionFunny face warpsEntertainment, viral content
BrandedLogo overlays, promotionalMarketing campaigns
3D ObjectsFace-tracked 3D elementsGaming, entertainment

Use Case Examples

1. Face Filter Design (Dog Ears and Nose)

curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "messages": [{"role": "user", "content": "Create an AR filter concept showing a young woman with realistic fluffy golden retriever dog ears on top of her head and a cute black dog nose overlay on her nose. Add subtle whisker dots on cheeks. The filter should look natural and integrate well with the face. Show front-facing selfie view."}],
    "model": "eachsense/beta",
    "stream": true,
    "mode": "max"
  }'

2. Beauty Filter Effect

curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "messages": [{"role": "user", "content": "Create a beauty AR filter effect showing before and after comparison. Show a person with the beauty filter applied: smooth skin, subtle face slimming, brightened eyes, soft glow effect, and enhanced jawline definition. Natural-looking enhancement, not overdone. Professional beauty app style."}],
    "model": "eachsense/beta",
    "stream": true,
    "mode": "max"
  }'

3. Makeup AR Filter

curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "messages": [{"role": "user", "content": "Create a virtual makeup AR filter concept. Show a woman with AR-applied makeup: bold red lipstick, smoky eye shadow in bronze and gold tones, defined eyebrows, subtle blush, and winged eyeliner. The makeup should look realistic as if professionally applied. Glamorous evening look. Front-facing portrait."}],
    "model": "eachsense/beta",
    "stream": true,
    "mode": "max"
  }'

4. Sunglasses and Accessories Filter

curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "messages": [{"role": "user", "content": "Create an AR try-on filter showing a person wearing virtual aviator sunglasses with gold frames and gradient lenses. Add a matching gold chain necklace and small hoop earrings. The accessories should look photorealistic and properly positioned on the face. Fashion e-commerce style visualization."}],
    "model": "eachsense/beta",
    "stream": true,
    "mode": "max"
  }'

5. Background Replacement Filter

curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "messages": [{"role": "user", "content": "Create an AR background replacement filter concept. Show a person in selfie view with their original background replaced by a tropical beach paradise - palm trees, turquoise water, golden sand, sunset sky. Clean edge detection around the person, natural lighting blend between subject and background."}],
    "model": "eachsense/beta",
    "stream": true,
    "mode": "max"
  }'

6. Face Distortion / Funny Filter

curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "messages": [{"role": "user", "content": "Create a funny face distortion AR filter. Show a person with comically enlarged eyes (anime style big), tiny mouth, and slightly enlarged forehead. Add sparkles around the eyes and a rainbow effect. Cute and comedic style like popular Snapchat filters. Exaggerated but still recognizable."}],
    "model": "eachsense/beta",
    "stream": true,
    "mode": "max"
  }'

7. Branded AR Filter (Logo Overlay)

curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "messages": [{"role": "user", "content": "Create a branded AR filter concept for a sports event. Show a person with face paint in team colors (blue and white stripes on cheeks), a virtual foam finger appearing next to their head, confetti particles falling, and a banner frame around the image that says GO TEAM. Celebratory sports fan atmosphere."}],
    "model": "eachsense/beta",
    "stream": true,
    "mode": "max"
  }'

8. Holiday Themed Filter

curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "messages": [{"role": "user", "content": "Create a Christmas holiday AR filter. Show a person wearing a virtual Santa hat, with a red Rudolph nose overlay, snowflakes falling around them, and a subtle frost effect on the edges of the frame. Add twinkling lights in the background. Festive and cheerful holiday spirit."}],
    "model": "eachsense/beta",
    "stream": true,
    "mode": "max"
  }'

9. Game-Style Filter (Among Us)

curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "messages": [{"role": "user", "content": "Create a gaming-inspired AR filter in the style of Among Us. Transform the person into a crewmate character - their face visible through a space helmet visor, wearing the iconic colored spacesuit (red), with a small pet companion floating nearby. Add a spaceship interior background. Fun and recognizable gaming aesthetic."}],
    "model": "eachsense/beta",
    "stream": true,
    "mode": "max"
  }'

10. 3D Object Face Tracking Filter

curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "messages": [{"role": "user", "content": "Create a 3D face tracking AR filter concept. Show a person with a 3D floating crown hovering above their head that follows head movement, 3D butterflies orbiting around them, and magical sparkle particles emanating from their eyes. The 3D elements should have realistic lighting and shadows. Fantasy magical theme."}],
    "model": "eachsense/beta",
    "stream": true,
    "mode": "max"
  }'

Best Practices

Face Filter Design

  • Natural Integration: Filters should blend seamlessly with facial features
  • Proper Tracking Points: Design around standard face landmark positions (eyes, nose, mouth, forehead)
  • Lighting Consistency: Match filter lighting with typical selfie lighting conditions
  • Scale Appropriateness: Elements should be proportional to face size

Platform Considerations

  • Instagram/TikTok: Vertical 9:16 format, mobile-first design
  • Snapchat: Consider lens studio capabilities and limitations
  • Video Calls: Subtle effects that work at lower resolutions

Performance Tips

  • Simplicity: Fewer elements = better real-time performance
  • Transparency: Use proper alpha channels for overlays
  • Animation: Consider how static designs translate to animated filters

Mode Selection

Ask your users before generating:

"Do you want fast & cheap, or high quality?"

ModeBest ForSpeedQuality
maxFinal filter concepts, client presentations, portfolio workSlowerHighest
ecoQuick drafts, concept exploration, brainstorming variationsFasterGood

Multi-Turn Creative Iteration

Use session_id to iterate on filter designs:

# Initial filter concept
curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "messages": [{"role": "user", "content": "Create a cat face AR filter with whiskers and cat ears"}],
    "model": "eachsense/beta",
    "stream": true,
    "session_id": "cat-filter-project"
  }'

# Iterate based on feedback
curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "messages": [{"role": "user", "content": "Make the ears more realistic and fluffy, add a pink nose, and include subtle face contouring to make cheeks more cat-like"}],
    "model": "eachsense/beta",
    "stream": true,
    "session_id": "cat-filter-project"
  }'

# Request variation
curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "messages": [{"role": "user", "content": "Create a variation with black cat features instead of tabby, and add glowing yellow eyes effect"}],
    "model": "eachsense/beta",
    "stream": true,
    "session_id": "cat-filter-project"
  }'

Filter Collection Generation

Generate multiple related filters for a campaign:

# Filter 1 - Dog
curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "messages": [{"role": "user", "content": "Create an AR filter with golden retriever dog ears and nose, cute and playful"}],
    "model": "eachsense/beta",
    "stream": true,
    "mode": "eco",
    "session_id": "animal-filter-pack"
  }'

# Filter 2 - Cat
curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "messages": [{"role": "user", "content": "Now create a matching cat filter with the same style - tabby cat ears and whiskers"}],
    "model": "eachsense/beta",
    "stream": true,
    "mode": "eco",
    "session_id": "animal-filter-pack"
  }'

# Filter 3 - Bunny
curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $EACHLABS_API_KEY" \
  -H "Accept: text/event-stream" \
  -d '{
    "messages": [{"role": "user", "content": "Complete the pack with a bunny filter - long floppy ears and a pink bunny nose"}],
    "model": "eachsense/beta",
    "stream": true,
    "mode": "eco",
    "session_id": "animal-filter-pack"
  }'

Error Handling

ErrorCauseSolution
Failed to create prediction: HTTP 422Insufficient balanceTop up at eachlabs.ai
Content policy violationProhibited contentAdjust prompt to comply with platform guidelines
TimeoutComplex generationSet client timeout to minimum 10 minutes

Prompt Tips for AR Filters

When creating AR filter concepts, include these details in your prompt:

  1. Filter Type: Specify the category (face filter, beauty, makeup, etc.)
  2. Elements: List all overlay elements needed (ears, nose, accessories)
  3. Style: Realistic, cartoonish, minimal, elaborate
  4. Positioning: Where elements appear relative to facial features
  5. Effects: Particles, glow, color grading, distortion
  6. Platform: Target platform if specific requirements exist
  7. Mood: Fun, glamorous, scary, festive, etc.

Example Prompt Structure

"Create a [filter type] AR filter showing [subject description].
Include [overlay elements] with [style/aesthetic].
Add [effects/particles] and [additional features].
[Platform/mood requirements]."

Related Skills

  • each-sense - Core API documentation
  • product-photo-generation - Product visualization
  • video-generation - Video content creation
  • image-generation - General image creation

平台分布

Codex

36.62%
按下载量换算272

Claude

27.08%
按下载量换算201

Cursor

17.81%
按下载量换算132

Gemini CLI

8.67%
按下载量换算64

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在提醒或未通过项,安装前请自行复核。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills