一键导入
dasheng-html-video-bridge
Use when Dasheng transwrite needs no-human or human-material talking-head videos rendered through the local html-video project.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Use when Dasheng transwrite needs no-human or human-material talking-head videos rendered through the local html-video project.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use when Dasheng Publish needs to route Xiaohongshu publishing or content access through API-first CLI, MCP, or browser fallback bridges.
Use when Dasheng Publish needs to prepare, upload, or package Bilibili video submissions from completed transwrite video channel packs.
Use when running Stage 3 Draft from approved Brief topics and producing Reasoning Sheets, standard drafts, and draft quality gates.
Use when the workflow enters Postmortem after Publish and published content needs to be evaluated with performance data, Publish Guard status, accuracy checks, and knowledge-base writeback suggestions.
Use when Dasheng draft or transwrite needs HTML Anything templates, editable article HTML, social cards, posters, or visual HTML references.
Use when running, resuming, auditing, or updating the Dasheng self-media workflow in this repo. This is the only formal orchestration entry and it governs the simplified canonical chain.
| name | dasheng-html-video-bridge |
| description | Use when Dasheng transwrite needs no-human or human-material talking-head videos rendered through the local html-video project. |
Use the external html-video repository as the default renderer for Dasheng video transwrite. This skill turns a confirmed Draft into an executable html-video project package; it does not research facts or rewrite the article thesis.
html-video is an external dependency, not vendored into this repo and not version-locked. Default path: /Volumes/PSSD/html-video; override with HTML_VIDEO_ROOT.
Check or install before first use:
python3 scripts/ensure_video_external_deps.py --dep html-video --mode check
python3 scripts/ensure_video_external_deps.py --dep html-video --mode install --install-node-deps
For Dasheng no-human explainer scenes, external html-video must also provide gsap and lottie-web; ensure_video_external_deps.py --install-node-deps checks and installs them in the external repo.
talking_head_video_manifest.jsonvideo_storyboard.jsontalking_head_script.mdResolve the root from HTML_VIDEO_ROOT or /Volumes/PSSD/html-video, then run:
HTML_VIDEO_ROOT="${HTML_VIDEO_ROOT:-/Volumes/PSSD/html-video}"
node "$HTML_VIDEO_ROOT/packages/cli/dist/bin.js" doctor --cwd "$HTML_VIDEO_ROOT"
node "$HTML_VIDEO_ROOT/packages/cli/dist/bin.js" search-templates --intent "<intent>" --aspect 9:16 --top 5 --cwd "$HTML_VIDEO_ROOT"
Prefer these templates for market commentary:
frame-liquid-bg-hero: hook/title/opening atmosphereframe-data-chart-nyt: one clear data comparisonframe-electric-studio: quote or conflict frameframe-light-leak-cinema: cinematic transitionframe-logo-outro: ending cardGenerate a project plan without mutating html-video:
python3 scripts/transwrite_html_video_bridge.py \
--video-manifest <talking_head_video_manifest.json>
Create and preview a real html-video project only when the user asks to render:
python3 scripts/transwrite_html_video_bridge.py \
--video-manifest <talking_head_video_manifest.json> \
--execute create
Render MP4:
python3 scripts/transwrite_html_video_bridge.py \
--video-manifest <talking_head_video_manifest.json> \
--execute render
Flow:
draft -> short script -> storyboard beats -> TTS audio -> html-video template vars -> preview -> MP4
Rules:
say for production voice unless explicitly accepted as a rough preview.mmx) as the default production provider for final narration, generated口播音频, background music, and AI image/video assets.Default voice render entry:
python3 scripts/render_html_anything_scene_pack_video.py \
--manifest <scene_pack_manifest.json> \
--output-dir <render_dir> \
--with-voice \
--voice-provider mmx \
--voice "Chinese (Mandarin)_Radio_Host" \
--mmx-model speech-2.8-hd \
--mmx-speed 1.08
Flow:
human audio/video -> Whisper/SRT -> beat alignment -> transparent/non-transparent HTML visual layer -> FFmpeg/html-video compose
Rules:
The bridge writes:
html_video_project_vars.jsonhtml_video_project_plan.jsonhtml_video_commands.shhtml_video_execution.json when executedtalking_head_video_manifest.json must point to html_video_project_plan.json.
rendered after an MP4 exists.html-video; install or update the external repo when the skill needs it.