技能 higgsfield-faceless-video

📦

higgsfield-faceless-video

Name: higgsfield-faceless-video
Author: higgsfield-ai

v0.14.0 內容修訂 r1 安全 ⚙️ 外部命令🌐 網路存取📁 檔案系統存取

Create Faceless Videos With Higgsfield

Producing narrated videos requires coordinated scripts, visuals, audio, and timing. This skill guides Higgsfield generation and local assembly with checks before delivery.

支援: Claude Codex Code(CC)

🥇 85 金級

Maintain this Skill? Claim it 🥇 85 金級

使用我的智慧代理安裝

將此請求複製給你的智慧代理，其中包含權威 Skill 頁面和清單。

智能體請求

Review the Skillstore skill "higgsfield-faceless-video" from https://skillstore.io/skills/higgsfield-ai-higgsfield-faceless-video.md and its manifest at https://skillstore.io/api/skills/higgsfield-ai-higgsfield-faceless-video/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.

你的智慧代理仍應顯示其計畫，並依安全政策請求必要確認。

Agent 可讀資源

當 AI Agent、爬蟲或腳本需要乾淨脈絡、而不是讀取完整頁面時，請使用這些連結。

Markdown 詳情 GET /skills/higgsfield-ai-higgsfield-faceless-video.md 已簽名 manifest GET /api/skills/higgsfield-ai-higgsfield-faceless-video/manifest 已簽名 lockfile GET /api/skills/higgsfield-ai-higgsfield-faceless-video/lockfile

測試它

正在使用「higgsfield-faceless-video」。 A 15-second English explainer with two illustrated scenes and no subtitles.

預期結果:

Expected deliverable: one assembled video with a 10-second opening block and a 5-second final block.
Supporting artifacts: a poster image and an assembly receipt with per-block speech timing.
Delivery note: measured duration, aspect ratio, language, subtitle status, and factual sources when applicable.

正在使用「higgsfield-faceless-video」。 Replace the middle 10-second range of a previously generated 30-second video.

預期結果:

Expected deliverable: a full-length review draft containing the replacement scene and surrounding source content.
Supporting artifact: an assembly receipt listing the edited range and measured output duration.
Review requirement: compare the edited range and surrounding content before publication.

安全審計

安全

v1 • 2026/10/28 開啟版本化報告

Of 365 static findings, 364 are false positives covering media tooling, temporary files, provenance checks, and documentation. One low-severity download risk remains because untrusted URLs and redirects can reach internal services. No evidence found of prompt injection, credential theft, or malicious execution intent in the reviewed files.

已掃描檔案

4,691

分析行數

複核項目

已忽略的誤報

能力複核項目 (1)

這些是真實的本機能力，對此技能而言可能是預期行為，因此需要複核，但不會按已確認的惡意行為計分。

低

scripts/assemble_blocks.sh:81

Fetch API call

fetch(){ if [[ "$1" == http://* || "$1" == https://* ]]; then curl -fsSL "$1" -o "$2"; else [[ -f "$

The downloader accepts arbitrary HTTP(S) destinations and follows redirects without address checks. An untrusted asset URL could trigger requests to internal services.

風險因素

⚙️ 外部命令 (50)

🌐 網路存取 (5)

scripts/assemble_blocks.sh:81 scripts/assemble_blocks.sh:81 scripts/tests/test_channel.py:39 scripts/tests/test_timeline.py:26 SKILL.md:5

📁 檔案系統存取 (44)

scripts/assemble_blocks.sh:76 scripts/assemble_blocks.sh:114 scripts/assemble_blocks.sh:121 scripts/assemble_blocks.sh:67 scripts/assemble_blocks.sh:202 scripts/assemble_blocks.sh:80 scripts/assemble_final.sh:83 scripts/assemble_final.sh:210 scripts/assemble_final.sh:60 scripts/assemble_final.sh:82 scripts/assemble_final.sh:111 scripts/assemble_partial_edit.sh:26 scripts/assemble_partial_edit.sh:39 scripts/assemble_partial_edit.sh:111 scripts/assemble_partial_edit.sh:112 scripts/assemble_partial_edit.sh:298 scripts/assemble_slides.sh:45 scripts/assemble_slides.sh:193 scripts/assemble_slides.sh:251 scripts/assemble_slides.sh:33 scripts/assemble_slides.sh:51 scripts/assemble_slides.sh:261 scripts/audio_to_captions.py:368 scripts/audio_to_captions.py:372 scripts/audio_to_captions.py:392 scripts/audio_to_captions.py:396 scripts/audio_to_captions.py:418 scripts/burn_captions.sh:48 scripts/burn_captions.sh:43 scripts/burn_captions.sh:47 scripts/burn_captions.sh:57 scripts/burn_captions.sh:61 scripts/burn_captions.sh:54 scripts/check_voice.sh:5 scripts/check_voice.sh:8 scripts/presenter_composite.sh:41 scripts/presenter_composite.sh:68 scripts/speech_metrics.sh:19 scripts/tests/test_assembly.py:6 scripts/tests/test_assembly.py:19 scripts/tests/test_channel.py:6 scripts/tests/test_channel.py:42 scripts/tests/test_modes.py:8 scripts/tests/test_modes.py:38

審計單位: codex

分享與引用此報告

分享版本化評估報告、中性徽章、嵌入卡片與引用。Skillstore 只報告證據，不判斷該 Skill 是否安全。

開啟版本化報告

複製報告連結

https://skillstore.io/skills/higgsfield-ai-higgsfield-faceless-video/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown 徽章

[![Skillstore security assessment](https://skillstore.io/badges/skills/higgsfield-ai-higgsfield-faceless-video/security.svg)](https://skillstore.io/skills/higgsfield-ai-higgsfield-faceless-video?utm_source=security_passport_badge)

HTML 徽章

<a href="https://skillstore.io/skills/higgsfield-ai-higgsfield-faceless-video?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/higgsfield-ai-higgsfield-faceless-video/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

嵌入卡片

<iframe src="https://skillstore.io/embed/skills/higgsfield-ai-higgsfield-faceless-video.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>

學術引用 (APA · BibTeX · CFF)

APA 引用

higgsfield-ai. (2026). higgsfield-faceless-video security audit report (audit version 1) [Author version 0.14.0]. Skillstore. https://skillstore.io/skills/higgsfield-ai-higgsfield-faceless-video/audits/1

BibTeX 引用

@techreport{higgsfield-ai-higgsfield-ai-higgsfield-faceless-video-2026,
  author = {higgsfield-ai},
  title = {higgsfield-faceless-video security audit report (audit version 1)},
  institution = {Skillstore},
  year = {2026},
  number = {1},
  url = {https://skillstore.io/skills/higgsfield-ai-higgsfield-faceless-video/audits/1},
  note = {Author version 0.14.0}
}

CITATION.cff

cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "higgsfield-faceless-video security audit report (audit version 1)"
version: "0.14.0"
type: report
authors:
  - name: "higgsfield-ai"
date-released: "2026-10-28"
url: "https://skillstore.io/skills/higgsfield-ai-higgsfield-faceless-video/audits/1"
identifiers:
  - type: other
    value: "skillstore:higgsfield-ai-higgsfield-faceless-video:audit:1"
    description: "Skillstore immutable audit report identifier"

Skillstore 評分

評分依據證據置信度: 中

架構

100

可維護性

內容

社群

100

規範符合度

你可以打造什麼

Illustrate a lesson

Create a narrated science explainer with diagram-led scenes, a consistent voice, and optional subtitles.

Produce a documentary episode

Plan sourced narration, recurring visual references, and ordered motion blocks for a channel episode.

Create an original children's song

Develop a supported-length song and matching illustrated scenes with consistent characters and a repeating chorus.

試試這些提示

Create a short explainer

Create a 30-second faceless explainer about how rain forms, in English and 16:9. Use illustrated diagrams, one narrator, and no music or subtitles. Show available voices and estimated costs before generation.

Build a narrated picture story

Create a 60-second picture story about a seed becoming a tree, in English and 9:16. Use storybook images, continuous narration, and subtitles. Verify local transcription prerequisites before paid generation and align scenes to the completed narration.

Produce a sourced channel episode

Create a 90-second channel-style motion episode about early navigation, in English and 16:9. Use reliable sources, a recurring ship reference, varied shots, and one narrator. Validate the script, measure narration takes, and assemble only after all checks pass. Retain job receipts for recovery.

Revise selected video ranges

Edit my previously generated 60-second faceless video using the supplied source and project assets. Replace only 10-20 seconds and 40-50 seconds with the supplied corrections. Preserve all other content and the existing style and voice. Validate patch ranges and replacement media, then assemble a full-length review draft. Do not upload or publish it.

最佳實務

Confirm duration, aspect ratio, voice, image roles, subtitle prerequisites, and generation costs before paid work.
Save job IDs, returned URLs, completed assets, scripts, and receipts in a persistent session directory.
Use trusted asset URLs and review the first scene, failed timing checks, captions, and final video before delivery.

避免

Speeding up narration or bypassing failed assembly checks to force a finished result.
Regenerating completed assets or submitting duplicate jobs while an earlier generation is still running.
Treating an incidental photo as presenter authorization or uploading a review draft without a request.

常見問題

Does the skill include free video generation?

No. It uses Higgsfield generation services and requires account access, available models, and credits. It checks estimated costs where supported.

Which local tools are required?

Core workflows use Bash, FFmpeg, ffprobe, Python 3, curl, and awk. Presenter processing also uses NumPy and Pillow.

Is narration sent to an external transcription service?

Caption transcription runs locally with faster-whisper and an existing model. Video, voice, and image generation still use Higgsfield services.

Can it create songs at any duration?

No. Song mode supports only 60 or 120 seconds. It does not support captions for singing.

Can it edit an existing faceless video?

Yes. It replaces selected, ordered, non-overlapping 10-second ranges and produces a full-length review draft. Other editing tasks are outside its scope.

Does it publish completed videos automatically?

No. Hosted delivery requires a request and uses the confirmed upload URL. Edited videos remain review drafts and must not be published automatically.

開發者詳細資訊

作者

higgsfield-ai

授權

MIT

作者版本

v0.14.0

Skillstore 修訂

儲存庫

https://github.com/higgsfield-ai/skills/tree/4d0757471a704dac8729ea3c2ef0f3f330e0d9b6/higgsfield-faceless-video

引用

311f8dc74431dfa24f919a5d1091d3d9619a0ddd

維護新鮮度

2026/10/28

使用量

0 次下載 · 0 次瀏覽

檔案結構

📁 references/

📄 captions.md

📄 channel-motion.md

📄 creative.md

📄 editing.md

📄 picture-story.md

📄 presenter.md

📄 production.md

📄 prompts.md

📄 song.md

📄 styles.md

📄 talking-characters.md

📁 scripts/

📄 assemble_blocks.sh

📄 assemble_final.sh

📄 assemble_partial_edit.sh

📄 assemble_slides.sh

📄 audio_to_captions.py

📄 bind_scene_frame_results.py

📄 build_scene_timeline.py

📄 burn_captions.sh

📄 check_voice.sh

📄 measure_narration_takes.py

📄 presenter_composite.sh

📄 speech_metrics.sh

📁 tests/

📄 test_assembly.py

📄 test_captions.py

📄 test_channel.py

📄 test_modes.py

📄 test_song.py

📄 test_timeline.py

📄 trim_static_head.sh

📄 validate_motion_script.py

📄 validate_song_prompt.py

📄 SKILL.md

higgsfield-ai 的更多技能

查看全部

📦

higgsfield-generate

生成 Higgsfield 圖像與影片

創意團隊需要可靠的模型路由來進行媒體生成。此技能會選擇 Higgsfield 工作流程、建立 CLI 指令，並傳回可用的媒體連結。

設計與創意查看

📦

higgsfield-product-photoshoot

使用 Higgsfield 建立產品攝影作品

產品團隊需要多種格式的活動圖片，而不必撰寫複雜的圖片提示詞。此技能會引導選擇、呼叫 Higgsfield，並傳回完成的圖片 URL。

設計與創意查看

📦

higgsfield-soul-id

訓練 Higgsfield Soul 角色

使用一次性提示詞來建立同一個人的一致 AI 圖像或影片很困難。此技能會引導使用已獲同意的照片進行 Higgsfield Soul 訓練，並回傳可重複使用的身分參考。

設計與創意查看

📦

higgsfield-marketplace-cards

建立 Marketplace 產品圖片組

Marketplace 刊登項目需要在主圖、資訊圖表、生活情境照和 A+ 模組之間保持一致的產品視覺。此技能會將清楚的刊登意圖導向 Higgsfield CLI 工作流程，產生可直接使用的圖片 URL。

設計與創意查看

📦

higgsfield-websites

建立 Higgsfield 網站與應用程式

推出精緻的網頁產品需要設定、設計、安全性與部署步驟。這項 skill 會引導 Claude、Codex 和 Claude Code 完成 Higgsfield CLI 網站與應用程式工作流程。

程式開發查看

📦

higgsfield-game-generation

Build Playable Browser Games with Higgsfield

Creating a coherent browser game requires coordinated design, assets, code, testing, and deployment. This skill guides that workflow through Higgsfield CLI and bundled asset tools.

程式開發查看

查看全部

📦

vibe-video

Create Music-Synchronized Educational Videos

由 immurray

Educational short videos require research, clear animation, and careful music timing. This skill coordinates agent work and local tools to produce narration-free vertical videos.

設計與創意查看

📦

video-sync

Sync Video With Shot Analysis

由 eternityspring

Shot notes are difficult to compare with footage when they live in separate files. This skill combines both into a synchronized video with an animated shot panel.

設計與創意查看

📦

brag

Create a Polished Launch Video

由 latent-spaces

Turning a project into a concise launch video requires product analysis, scripting, animation, audio, and rendering. Brag coordinates that workflow with Hyperframes and delivers a video, poster, storyboard, and share copy.

設計與創意查看

📦

generating-storyboards

Create Consistent Video Storyboards

由 supercmohq

Video concepts can lose continuity when clips are planned separately. This skill creates sequential storyboard sheets that preserve characters, products, settings, and visual states.

設計與創意查看

📦

ponyflash

使用 PonyFlash 生成與編輯 AI 媒體

由 ponyflash

AI 媒體工作通常橫跨獨立的生成、編輯與字幕工具。PonyFlash 結合 SDK 指引與 FFmpeg 工作流程，支援影像、影片、語音、音樂與匯出。

設計與創意查看

📦

general-video

建立自訂 HyperFrames 影片

由 heygen-com

自訂影片需求通常需要在設計、動態、音訊及驗證方面具備結構化流程。此技能會透過 HyperFrames 工作流程安排工作，以確保製作一致性。

設計與創意查看

higgsfield-faceless-video

使用我的智慧代理安裝

Agent 可讀資源

測試它

安全審計

風險因素

複製報告連結

Markdown 徽章

HTML 徽章

嵌入卡片

APA 引用

BibTeX 引用

CITATION.cff

Skillstore 評分

你可以打造什麼

Illustrate a lesson

Produce a documentary episode

Create an original children's song

試試這些提示

最佳實務

避免

常見問題

開發者詳細資訊

higgsfield-ai 的更多技能

higgsfield-generate

higgsfield-product-photoshoot

higgsfield-soul-id

higgsfield-marketplace-cards

higgsfield-websites

higgsfield-game-generation

相關技能

vibe-video

video-sync

brag

generating-storyboards

ponyflash

general-video