Generate and edit Draw.io, Mermaid, and Excalidraw diagrams from natural language using a structured JSON spec.
Design & media
sn-motion-html
Try it从主题策划到 AI 图片、视频和滚动网页的一体化制作
What it does
Create a reusable project, not a one-off HTML file. Keep subject matter, chapter count, visual style, interface treatment, copy placement, language, image model, video model, clip duration, and hosting path configurable.
The skill document
SN Motion HTML
Create a reusable project, not a one-off HTML file. Keep subject matter, chapter count, visual style, interface treatment, copy placement, language, image model, video model, clip duration, and hosting path configurable.
The defining experience is one continuous camera journey controlled by page progress. Preserve a single fixed media stage and one ordered timeline from the first scene through every connector to the final scene. Differentiate projects through transparent copy composition, header treatment, type, navigation, and scene-specific copy placement—not by replacing the continuous stage with separate article media blocks.
Route the work
- Resolve the story goal, audience, factual fidelity, chapter scale, visual language, copy language, target devices, sound policy, and available API credentials. Make reasonable defaults only for low-risk implementation details.
- For factual or source-based stories, research before outlining and record sources separately from display copy.
- Produce a chapter blueprint, shot/connector plan, and character/object bible before generating the full media set.
- Detect the image-generation capability available in the current environment before generating any stills. The capability may be a callable image tool, a skill that exposes or instructs an image workflow, or both. Use the available capability when present; if no suitable capability is available, report that blocker to the user before continuing past the still-generation stage.
- Prepare a review package containing the proposed visual style and the content/shot arrangement. When an image capability is available, generate only the small reference/gate set needed for review.
- Obtain the user's explicit approval at both approval gates below. After approval, generate the complete still set and inspect it before any paid video batch.
- Dry-run the video manifest and representative clip. Obtain explicit authorization immediately before the first paid video API request, then generate scene clips and connector clips in dependency-aware parallel phases.
- Choose or design an interface treatment around the continuous stage, assemble from structured content, then verify media and browser behavior.
Required user approval gates
Do not silently infer approval from an unanswered message or from a prior general instruction. If the user requests changes, revise the package and present the affected gate again.
- Visual style approval: present the style preamble and identity blocks, including medium, palette, materials, lighting, camera grammar, composition, negative constraints, and recurring character/product treatment. When image generation is available, include a small reference/gate set (normally opening, middle, and ending stills). Do not generate the complete still set until the user explicitly approves the visual direction.
- Content and shot-plan approval: present the chapter order, each chapter's dominant visual event and display copy, scene framing/camera intent, connector start and destination, and per-scene copy position. Do not generate the complete still set or write the final video manifest until the user explicitly approves this arrangement.
These are separate decisions: style approval determines how the world looks; content and shot-plan approval determines what is shown and how the camera moves through it. The later paid-video approval is an authorization gate, not a substitute for either creative approval.
Read references/workflow.md for the full production sequence. Read only the additional references needed for the current stage:
- references/schemas.md when creating manifests or content JSON.
- references/prompting.md when generating character references, scenes, or motion prompts.
- references/style-presets.md when choosing or adapting anime, cinematic, or CGI production styles.
- references/seedance.md before configuring or calling the video API.
- references/frontend.md before selecting or adapting the interface treatment.
Start from the template
Run:
python scripts/init_project.py /absolute/output/path --title "Story title" --style cinematic --ui folio
The initializer copies assets/template/ and creates the media directories. Replace the example story and manifests; do not leave placeholder subject matter in the final project.
Choose --style anime, --style cinematic, or --style cgi. Separately choose --ui folio, --ui caption, or --ui graphic. Style controls the generated world; UI controls header, transparent copy treatment, navigation, and type composition. Per-chapter copyPosition may be left, right, bottom, or center when the focal subject requires it.
All UI presets use the same continuous-shot runtime. Adapt the visual chrome materially for the subject, but do not interrupt the camera chain or turn chapters into separate media cards.
It also copies the reusable scripts into the new project. From the project root, the common commands are:
python scripts/seedance_pipeline.py . plan
python scripts/seedance_pipeline.py . all --workers 4
python scripts/verify_media.py .
python scripts/contact_sheet.py .
python scripts/serve_project.py . --port 8080
Production invariants
- Keep narrative data in
content/story.json; do not hardcode chapter copy into the runtime. - Keep each chapter's complete, concise copy visible together. Do not rotate paragraphs within one clip unless explicitly requested.
- Maintain a single style preamble and immutable identity blocks for recurring characters or products.
- Require explicit user approval of the visual style and the content/shot plan before generating the complete still set or final video manifest. Run a small still-image consistency gate before generating the full set. The gate requires an available image-generation capability (callable tool, applicable skill workflow, or both); do not silently substitute placeholder stills when that capability is absent.
- Use one fixed full-viewport media stage. Ordered scene and connector clips together form one virtual continuous take.
- Scene clips and every adjacent connector are required unless the user explicitly accepts a visible cut.
- Keep one complete, static, transparent copy block per scene clip. Connector clips carry no changing paragraph sequence.
- Copy articles have no solid card fill by default. Use localized edge-free gradients, text shadow, or fine rules for contrast so the words remain visually integrated with the scene.
- Dives may run in parallel. Connectors must wait until all adjacent dive boundary frames exist, then may run in parallel.
- Immediately download provider results. Normalize all final clips to identical codec, dimensions, frame rate, exact duration, no audio unless requested, short GOP, and
faststart. - Use project-local
.env, exclude it from version control, and ensure a static server cannot expose it. - Never print secrets or embed them in frontend files. Do not infer authorization for paid generation; if it is not explicit, stop after the dry run. Creative approval does not imply paid-generation authorization.
- Bound automated retries. Treat authentication, permission, billing, and model-activation errors as non-retryable.
- Preserve still posters and a reduced-motion path so the story remains usable when video loading or motion is unavailable.
Validation
Before delivery:
- Validate JSON, JavaScript, shell, and Python syntax.
- Run
scripts/verify_media.py PROJECT_ROOT; require the declared count, dimensions, frame rate, codec, and exact duration. - Test a fresh browser session, reverse and forward progress, chapter navigation, transition boundaries, missing-video fallback, and reduced motion.
- Inspect a midpoint contact sheet. Prioritize correct identities, key objects, and scene meaning over incidental generative details unless the user asks for stricter fidelity.
- Report the preview URL, project path, media directories, model configuration, QA result, and any remaining external blocker.
Related skills
Stores durable facts in a categorized, plain-markdown vault on disk, alongside your agent's built-in memory.
Join a video meeting as an AI bot with voice, avatar, and screenshare across four operating modes.
Find why your productivity system keeps failing, then apply the smallest fix — capacity math, bottleneck routing, durable local notes.
Fetch raw ad creative, app, ranking, and revenue data from AdMapix as structured JSON.
Site audit, content writing, and competitor analysis for organic search rankings.
More from sensenova-skills
Browse all skills用于用户请求深度研究、系统性研究、竞品分析、方案对比、趋势分析或事实核查时。**遇到以下任一情况就主动使用本 skill,不要自行搜几条就回答**:①用户出现触发词:深度研究 / 深度调研 / 深入研究 / 全面研究 / 系统研究 / 调研 / 调查 / 尽调 / 行业研究 / 市场研究 / 竞品分析 / 政策研究 / 技术研究 / 趋势研究 / 事实核查 / 写一份研究报告 / 调研报告 / 深度报告 / research / deep research;②请求需要跨多来源取证、多维度对比、交叉验证才能给出可靠结论;③用户要求产出报告、白皮书、行业分析或尽调文档;④话题涉及最新政策/市场/产品/价格/法规,需要系统核查。无核验要求的简单常识问答不使用。模糊或宽泛的"研究/了解一下 X"也优先触发。仅不用于:一句话摘要、已给定单一来源的整理、纯文字润色改写。
Generates professional infographics with various layout types and visual styles. Analyzes content, recommends layout and style, and generates publication-ready infographics. Use when user asks to create "infographic", "信息图", "visual summary", or "可视化".
Open the AI PPT editing WebUI for previously generated SenseNova HTML slides. Use when the user asks to preview, inspect, edit, or open an existing/generated PPT deck in the WebUI/workbench without regenerating slides. Accepts an explicit deck_dir or auto-detects the newest local ppt_decks entry containing HTML pages. Does not run style/outline/page-html/export generation.
多来源深度调研、系统分析、事实核查与完整研究报告交付
Base-layer skill for the SenseNova-Skills project, providing low-level APIs for image generation, recognition (VLM), and text optimization (LLM). This skill does not preprocess inputs; it only calls backend services and returns results. This skill is not user-facing and is intended for upper-layer skills only.
Standard and fast PPT pipeline. All LLM / VLM / T2I calls are wrapped in a single CLI entry (scripts/run_stage.py). The main agent's job is simple: emit ONE shell command per stage, never write loops, never write prompts. Standard mode plans thoroughly with a three-sample deck preview checkpoint (three concatenated deck images plus a preview URL), web research, image search, and user-selected final output format (PPTX or PDF) for polished, delivery-ready presentations. Fast mode builds a complete draft immediately with autonomous decisions, then provides structured refinement suggestions so the user can iterate quickly. Supports AI-generated infographics (U1) for diagrams and flowcharts, web image search (Serper) for real photos, and ECharts for data charts.