Communitygithub.com

Higgsfield 灵魂角色训练

基于人脸训练个性化的身份忠实模型,用于 Higgsfield 的图像和视频生成。

Higgsfield 灵魂角色训练 是什么?

此技能用于在 Higgsfield 平台上训练一个“灵魂角色”(Soul Character),即一个基于用户提供照片的面部身份忠实模型。训练是一次性过程,完成后会返回一个 reference_id,可在后续的图像和视频生成任务中通过 `--soul-id` 参数重复使用。支持的模型包括 `text2image_soul_v2`(图像生成)和 `soul_cinema_studio`(视频生成)。该技能适合需要创建个人数字分身、定制化头像或用于视频生成的场景,但不适用于一次性换脸(应使用 `higgsfield-generate` 配合 `--image` 参数)或非照片类角色。训练前需确保用户已安装 Higgsfield CLI、完成登录认证,并且拥有付费订阅(Basic 以上)。工作流程包括获取角色名称、收集 5–20 张高质量面部照片、选择合适的模型变体、提交训练任务、静默等待完成,最后提供 reference_id。

兼容平台~Claude Code~Codex CLI~Cursor
npx skills add https://github.com/adamroy13/ai-influencer-promo-machine/tree/HEAD/.agents/skills/higgsfield-soul-id

在你喜欢的 AI 中提问

打开一个已预加载此 Agent Skill 的新对话。

文档

Higgsfield 灵魂角色训练 是做什么的?

训练一个面部身份忠实模型。可在所有支持灵魂角色的生成任务中重复使用。

第零步 —— 初始化

在执行任何其他命令前:

  1. 如果 higgsfield 不在 $PATH 中,请安装:
    curl -fsSL https://raw.githubusercontent.com/higgsfield-ai/cli/main/install.sh | sh
    
  2. 如果 higgsfield account status 返回 Session expired / Not authenticated,请让用户运行 higgsfield auth login(交互式)并等待确认。
  3. 灵魂角色训练需要付费订阅(Basic 及以上)。如果 higgsfield account status 显示为免费计划,请在提交前告知用户。

用户体验规则

  1. 保持简洁。不要在聊天中显示原始 ID。只需说“灵魂角色就绪”并引用名称。
  2. 检测用户语言并以此语言回复。CLI 参数保持英文。
  3. 只询问最少的信息:名称 + 照片。选择一个合理的模型变体。
  4. 静默轮询——训练需要几分钟。不要重复更新状态。

工作流程

  1. 获取名称。 一个单词,用于后续引用。如果缺失则询问。
  2. 获取照片。 5–20 张面部照片,需包含不同角度和光照条件。本地路径或已上传的 ID 均可 —— --image 参数接受两者。
  3. 选择模型变体。
    • --soul-2 —— 用于图像生成(默认)
    • --soul-cinematic —— 用于电影/视频工作 根据用户声明的下游用途选择。默认使用 --soul-2。
  4. 提交任务。
    higgsfield soul-id create --name "<名称>" --soul-2 --image ./photo1.png --image ./photo2.png ...
    higgsfield soul-id create --name "<名称>" --soul-2 --image <upload_id> --image <upload_id> ...
    
    CLI 会自动上传本地路径。捕获返回的参考 ID。
  5. 等待。 执行 higgsfield soul-id wait <id>。静默模式。默认超时 30 分钟。
  6. 交付。 “灵魂角色 <名称> 就绪。在生成时使用 --soul-id <id>。”

使用灵魂角色

训练完成后,将参数传递给 higgsfield-generate:

higgsfield generate create text2image_soul_v2 --prompt "..." --soul-id <引用ID> --quality 2k --wait
higgsfield generate create soul_cinematic --prompt "..." --soul-id <引用ID> --quality 2k --wait

列出现有灵魂角色

higgsfield soul-id list                   # 列出所有引用
higgsfield soul-id get <id>               # 按 ID 查询单个

错误处理

  • Minimum Basic plan required —— 用户处于免费计划;请告知。
  • Training failed —— 检查照片质量(至少 5 张独特的面部照片,光线充足)。
  • Session expired → 执行 higgsfield auth login。

参考文档

  • references/photo-guide.md —— 哪些照片效果最好
  • references/troubleshooting.md —— 常见训练失败原因

Individual skills in this repo

This repo contains 4 individual skills — each has its own dedicated page.

adamroy13/ai-influencer-promo-machine

A 5-stage AI content pipeline using Higgsfield MCP, focused on viral UGC content split evenly across 5 formats: UGC Entertainment (challenges), Street Interview, Unboxing, Product Review, ASMR. Button-driven UX — clicks, no typed commands. Stage 1: probe Marketing Studio capability, research trends in the product's niche on Instagram, TikTok, YouTube, and turn them into prompts mapped to the 5 formats with picklist hooks. Stage 2: produce one HTML video content plan; user sets the count, the skill divides evenly across the 5 formats. Stage 3: generate videos one format batch at a time, then auto-generate the image asset pack via GPT Image 2.0. Stage 4: schedule to Meta Ads via Meta MCP. Stage 5: render a cost comparison — Higgsfield credits + USD spent vs traditional production cost. Trigger on a product image + content request, or phrases like 'create a campaign', 'build a content plan', 'make videos in Higgsfield', 'generate the image pack', 'schedule to Meta', 'run the content pipeline'.

adamroy13/ai-influencer-promo-machine

Generate images/videos via Higgsfield AI. Default: GPT Image 2 for images/design/text, Seedance 2.0 for video, Nano Banana 2/Pro for character/reference image work, Marketing Studio for ads with avatars/products/hooks, settings, plus Soul V2/Cinema/Cast/Location and Kling 3.0. Use when: "generate an image", "make a video", "animate this photo", "image-to-video", "edit/stylize/remix this image", "produce a clip", "create an ad", "make a UGC video", "product demo", "unboxing", "brand video", "presenter video", "import product from URL", "create avatar for ad", or "analyze video virality". Supports image-to-image, image-to-video, references, job/upload IDs, and Marketing Studio. Chain with higgsfield-soul-id for face/identity consistency. Virality Predictor (`brain_activity`) analyzes video virality: hook strength, attention, retention, distraction risk, and creative score. NOT for: Soul Character training (use higgsfield-soul-id), product photoshoots, marketplace listing cards, text/chat/TTS tasks.

adamroy13/ai-influencer-promo-machine

Generate marketplace product image cards through Higgsfield: compliant main image, secondary product images, and A+ style content modules. Use when the user asks for marketplace listing images, product detail cards, secondary product images, product infographics, lifestyle listing shots, A+ style content, marketplace image sets, or sales-ready product visuals. Backend owns marketplace compliance references and prompt templates; this skill only routes user intent to the CLI. NOT for generic brand product photography without marketplace/listing context (use higgsfield-product-photoshoot), video generation or UGC ads (use higgsfield-generate), or Soul Character training (use higgsfield-soul-id).

adamroy13/ai-influencer-promo-machine

Generate brand-quality product images through Higgsfield product-photoshoot prompt enhancement on GPT Image 2 / gpt_image_2. Entry point for professional brand/product visuals. Use when: "product photo", "studio shot", "lifestyle image", "Pinterest pin", "hero/banner", "carousel", "ad creative", "Meta ads", "virtual try-on", "model wearing", "person holding product", "closeup with hands", "levitating/floating/splash product", "CGI/surreal product", "restyle", "seasonal/aesthetic variation", or any product, brand, or paid-social creative. Modes: product_shot, lifestyle_scene, closeup_product_with_person, moodboard_pin, hero_banner, social_carousel, ad_creative_pack, virtual_model_tryout, conceptual_product, restyle. Backend assembles the final prompt; never freehand it. NOT for: no-product text-to-image (use higgsfield-generate), branded avatar video (use higgsfield-generate Marketing Studio), marketplace listing cards (use higgsfield-marketplace-cards), Soul Character training (use higgsfield-soul-id).

相关技能