dsh-infinite_canvas

Infinite Canvas for DeepSeek Harness

A node-based short-drama studio inside DeepSeek Harness. Start from a one-line idea, walk out with a storyboarded, voiced, subtitled video cut — right on an infinite canvas.

English | 简体中文

What is this?

Infinite Canvas is a DeepSeek Harness plugin that turns the web GUI's sidebar into a full creation studio. Instead of chat round-trips, you work the way directors do: nodes on an infinite canvas, wired into a pipeline — script, storyboard, character & scene assets, image generation, video generation, and a final cut with transitions, subtitles and dubbing.

Everything runs locally in your Harness. Bring your own generation APIs, or use the built-in demo mode to learn the whole flow with zero configuration.

How it works

flowchart LR
    A[Script node<br>AI screenplay] --> B[Storyboard node<br>AI shot list]
    B --> C[Expand to canvas<br>1 shot = 1 node group]
    C --> D[Image gen<br>multi-reference]
    D --> E[Video gen<br>first frame + refs]
    E --> F[Final cut<br>transitions · subtitles · voices]
    F --> G[Export<br>WebM preview / FilmCraft MP4]

Highlights

Consistent characters, scene after scene

A per-project series bible (style baseline + character/scene/prop cards) is injected into every AI call, so wardrobes, locations and props stay coherent across shots and episodes. After each storyboard run, a local consistency check flags any name that isn't registered yet.

References that actually bind

Reference images connect as an ordered, visible list — [图1] 角色：Lin Ning, [图2] 场景：The Old House — and that exact order is what reaches the model. Multi-character frames stay unambiguous, and Volcano Engine Seedance's native [图N] convention is supported end to end (first frame, up to 4 reference images, official task API).

From shots to a watchable cut

The Final Cut node stitches shot videos in order — crossfade or black transitions, burned-in subtitles, original audio and TTS voices mixed in — and exports a preview cut instantly. For real post-production, one click writes a FCP7 XML timeline + SRT to disk and hands the render to ArtCraft (FilmCraft) for a true MP4.

An editor-grade canvas

Pan, zoom, marquee select, alignment guides, groups that fold, a minimap, 60-step undo/redo, per-node media history (last 8 results), mask-based inpainting, and middle-click everything — the canvas part should feel like an editor, not a chat sidebar.

Bring your own models

Text models come from Harness itself (no extra keys). Image, video and TTS backends are multi-config — define as many endpoints as you want, mark a default, and pick per node. No backend configured? Demo mode synthesizes placeholder images and animated videos locally so the entire pipeline is testable offline.

Quick start

1. Install the plugin (bundle folder) and restart Harness once.
2. Click the Infinite Canvas button in the left sidebar.
3. Add a Script node, type a logline, hit AI write script.
4. Connect it to a Storyboard node → AI break into shots → Expand to canvas.
5. Drop character images onto the placeholder asset nodes, hit Run workflow, then open a Final Cut node and export.

Without any generation API configured, every step runs in demo mode — the full pipeline works offline.

Generation backends

Kind	Protocol	Notes
Text	Harness LLM service	Script & storyboard writers; pick any configured model, or override per node
Image	OpenAI-compatible /images/generations, multi-reference via /images/edits	Size (1K/2K/4K) × ratio; mask inpainting via edits + PNG mask
Video	Generic submit+poll, or Volcano Seedance official tasks	duration 4–10s, 480P/720P/1080P, ratio, first frame, up to 4 reference images
TTS	OpenAI-compatible /audio/speech	Per-shot voices mixed into the final cut

ArtCraft bridge

When the companion dsh-artcraft plugin is installed, the canvas talks to the local PhotoCraft / FilmCraft / EffectCraft toolchain over same-origin HTTP: assets sync to a per-project data folder, images get headless PhotoCraft grading (single or batch), and final cuts export as FCP7 XML + SRT for FilmCraft to render as MP4. Status lights, graceful degradation and a full contract live in docs/artcraft-handoff.md.

Publishing (maintainers)

Testing funnel, beta tags, pack validation and the full release runbook live in docs/release.md.

Publishing runs through npm Trusted Publishing (OIDC) — no npm token exists anywhere:

• Release flow: create a GitHub Release (tag v0.17.0, …) → Actions publishes that version automatically.
• Manual flow: repo → Actions → publish → Run workflow → type a version.
• The publish.yml workflow resolves the version from the release tag (or your input) and publishes with a provenance badge.

Data & privacy

• Canvas documents, projects, settings and the series bible live in browser localStorage (dsh-ic:*); assets live in IndexedDB (dsh-infinite-canvas) — or on disk once a data root is set in Settings (per-project folders under the configured root, served by the plugin backend with HTTP Range support for video seeking).
• AI text generation runs through the Harness host process (your configured models, no extra keys).
• Image/video/TTS calls go directly from the browser to the endpoints you configure — mind CORS and key handling.
• Nothing is uploaded anywhere by the plugin itself.

Requirements & limitations

• DeepSeek Harness 0.2.0-rc.2+; demo video export needs MediaRecorder (Chromium).
• The preview cut is real-time recorded WebM; for MP4 masters with multi-track editing, pair with ArtCraft FilmCraft or any NLE.
• Image edits/mask inpainting requires a model that supports the endpoint (e.g. gpt-image-1).
• Host-side plugin updates need one Harness restart; the web half hot-reloads.

License

Apache-2.0