@wji.ai/cli 0.8.1 → 0.8.3

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@wji.ai/cli",
-  "version": "0.8.1",
+  "version": "0.8.3",
   "description": "Command line interface for the WJI AI platform",
   "license": "UNLICENSED",
   "author": "WJI AI",
@@ -28,18 +28,19 @@
     "README.md"
   ],
   "engines": {
-    "node": ">=20"
+    "node": ">=22.23.4"
+  },
+  "scripts": {
+    "build": "tsup && chmod +x dist/index.js",
+    "dev": "tsup --watch",
+    "typecheck": "tsc -p tsconfig.json --noEmit",
+    "start": "node dist/index.js",
+    "prepublishOnly": "pnpm run build"
   },
   "dependencies": {
     "commander": "^12.1.0"
   },
   "publishConfig": {
     "access": "public"
-  },
-  "scripts": {
-    "build": "tsup && chmod +x dist/index.js",
-    "dev": "tsup --watch",
-    "typecheck": "tsc -p tsconfig.json --noEmit",
-    "start": "node dist/index.js"
   }
-}
+}

package/skill/short-drama/SKILL.md

@@ -59,7 +59,11 @@ ending), change the guideline first, then the episodes.
    古装 / 年代 / 英文美式 / 英文英式 — it decides how the lines are written and the accent the video
    model speaks), ratio (default 9:16 for short, 16:9 for medium and movie). Offer the defaults you will use unless
    told otherwise: `video: h3` (MiniMax H3 ref), `style: realistic`, **`references: sheets`** (no
-   storyboard images) and **`videoDuration: auto`**. Do not start Stage 1 without the first two.
+   storyboard images) and **`videoDuration: auto`**. That message is a gate. If the user left
+   something out — no video model, no episode count, no genre, no 语体 — ask for it, or name the
+   default you would use, in the same message. Do not `init`, and do not start Stage 1, until they
+   confirm the set. A default is a proposal, not a decision. A resumed `project.json` that already
+   has these params is not asked again unless a required field is empty.
 
    **The video mode decides how the plot is cut up.** A segment is one video generation, so its
    length is whatever the provider accepts — and that differs by a factor of two:
@@ -132,7 +136,9 @@ ending), change the guideline first, then the episodes.
    `params.violence` is `blood` (default) or `implied`.
 4. **Locate or create the project directory** (the user's choice, else next to the source):
    ```bash
-   node {baseDir}/scripts/short-drama.mjs init <dir> --title "<title>" --episodes N --minutes M --genre "<genre>" --register 台湾国语 [--adapt 抽核] [--ratio 9:16] [--video h3|seedance|seedance2|seedance2fast|wan] [--resolution 480P|…] [--accent "台湾口音国语"] [--style photoreal|realistic|ghibli] [--references sheets|keyframes] [--duration auto|fixed] [--look "<lens phrase>"] [--cinematography low-key|sweet-romance|period-nostalgia|costume-idol|costume-dark|scifi-hard|scifi-soft|none] [--action gun-fu|hk-kungfu|wuxia-fantasy|historical-realist|bullet-time|superhero|street-brawl|car-chase|none] [--violence blood|implied] [--grammar drama|action|faceslap|comedy] [--comedy slapstick|deadpan|romcom|none] [--composition standard|art-first] [--story revenge,sweet,…] [--market mainland|overseas] [--format short|medium|movie] [--lang zh]
+   node {baseDir}/scripts/short-drama.mjs init <dir> --title "<title>" --episodes N --minutes M --genre "<genre>" --register 台湾国语 [--adapt 抽核] [--ratio 9:16] [--video h3|seedance|seedance2|seedance2fast|wan] [--resolution 480P|…] [--accent "台湾口音国语"] [--style photoreal|realistic|ghibli] [--references sheets|keyframes] [--duration auto|fixed] [--look "<lens phrase>"] [--cinematography low-key|sweet-romance|period-nostalgia|costume-idol|costume-dark|scifi-hard|scifi-soft|none] [--action gun-fu|hk-kungfu|wuxia-fantasy|historical-realist|bullet-time|superhero|street-brawl|car-chase|none] [--violence blood|implied] [--grammar drama|action|faceslap|comedy] [--comedy slapstick|deadpan|romcom|none] [--composition standard|art-first] [--story revenge,sweet,…] [--market mainland|overseas] [--format short|medium|movie] [--lang zh] [--image-lang en|zh]
+   # --image-lang zh: character / scene / prop image prompts in Chinese (the default is English)
+   # --foreground low|medium|high: how often cuts shoot past a foreground (default medium; high for 唯美 / 美感优先)
    # params.camera (body / lens / aperture) is edited in project.json
    ```
    An existing `project.json` means a resumed run: `validate` and `manifest` tell you where it stands.
@@ -161,27 +167,33 @@ write JSON → validate --stage <stage> → fix until green → render → sync
 
 ## Production dashboard
 
-Generate the dashboard whenever the user wants to review the project, and refresh it after a
-stage or episode is locked:
+The dashboard is one static file, `{baseDir}/dashboard.html`. It does not bake `project.json`,
+and it does not download images. **Every project keeps its own copy in the same folder as
+`project.json`.** `init` copies it. On any later session, if that file is missing, copy it
+again — do not overwrite a copy that is already there:
 
 ```bash
 node {baseDir}/scripts/short-drama.mjs dashboard {project}
-# custom location, and a folder of downloaded WJI images:
-node {baseDir}/scripts/short-drama.mjs dashboard {project} \
-  --output {project}/dashboard.html --assets {project}/assets --force
 ```
 
-The output is one static HTML file: production totals and gate health, brief and guideline,
-episode directory, character bibles, scene / prop sheets, scripts, cut timing and storyboard
-plans. It does not call WJI and costs no credits. Every character, scene, prop and storyboard
-slot keeps an image placeholder until a picture or video is bound. A bound value may be a WJI
-server URL (`https://…` image or video) or a local file: set `image.url` / `video.url`, or
-`wji.assets.sheet|headshot|keyframe|video`, to that URL or to a path relative to the project
-(or an absolute path). Files whose name contains the entity or asset id are also picked up from
-`{project}/assets`, `{project}/output`, the project tree, or `--assets <dir>`. A video URL or
-`.mp4` / `.webm` / `.mov` file renders as a player; if the address fails to load, the
-placeholder stays. Open `dashboard.html` in a browser. Do not paste the HTML into a dataset or
-upload it as an asset.
+That command checks `{project}/dashboard.html`. Missing → copy. Already there → leave it
+(`--force` replaces it with the skill's current page). Open the copy through a local static
+server so it can `fetch` `./project.json` and `episodes/E01.json`. Refreshing re-reads the JSON.
+
+Pictures and videos stay as asset ids on `wji.assets.sheet|headshot|keyframe|poster|video`
+(and `lighting:` / `state:` keys). The page asks `POST /assets/check-status` for those ids
+and uses the returned `url` — the same file address as `GET /assets/list/:id` (`wji asset get`).
+Paste the bearer from `wji auth token` once; the browser then loads the CDN url. Do not run
+`wji asset download` for the dashboard. An `https` url already stored on `image.url` /
+`video.url` is used as-is. A relative file that is already in the chosen folder is shown from
+that file. Anything else stays a placeholder. Video urls render as a player.
+
+The left nav scrolls the page itself (the section links call `scrollTo` on the content column).
+Bound WJI ids become links to `https://app.wji.ai/app/workspace/<workspaceId>`:
+a canvas is `?app=InfiniteCanvas&canvas=<id>` (and `&node=<id>` when a node is bound);
+a dataset is `?app=Dataset&dataset=<id>`.
+
+Do not paste the HTML into a dataset or upload it as an asset.
 
 ## Stage 1 — Outline
 
@@ -286,14 +298,20 @@ Read `{baseDir}/reference/stage-storyboard.md`.
 Every cut is a set-up, so besides its beats and framing it carries the **lens** (`focal`, or let the
 shot size pick it), the **key light relative to the camera** (`light.direction` from the vocabulary
 plus `light.note` saying what it picks out, what fills the shadow side and how it changes over the
-cut) and the **foreground** to throw out of focus. All three are gated and all three go into the
+cut) and the **foreground** if the story gives one a reason (`params.foregroundLevel`: `low` / `medium` / `high`). All three are gated and all three go into the
 prompt, alongside the camera package and each speaker's delivery.
 
+**Everyone on screen is alive** (gate `idle-motion`, sheets mode): a model freezes whoever the prompt gives no
+  motion. In a cut with two or more visible people, all but the one carrying the action get a small
+  motion in `idle` (`"still"` when keeping still is the performance). See `craft-motion.md`.
+
+**Two people talking** (gate `two-person-framing`, sheets mode): a two-shot once, then over-the-shoulder (`ots`) for the exchange, a clean single for the punchline. A single names only its person; every look states a screen direction (`eyeline`); the sides stay locked. See `reference/craft-shots.md`.
+
 **Every cut is lit and composed in the preset's look.** Read the preset's guide
 (`{baseDir}/reference/cinematography-<preset>.md`) before the first storyboard. Name the source
 in `light.source`, stage the person against it, and choose the camera side that gives the
 preset's preferred direction (the table in `cinematography.md`). Compose as the guide says,
-shoot past a foreground, and stay within the preset's stop (T2.8; T2 for `sweet-romance` and `costume-idol`, which
+and stay within the preset's stop (T2.8; T2 for `sweet-romance` and `costume-idol`, which
 start on a T1.3 package and name their bokeh; T4 for `period-nostalgia`). Both costume presets (`costume-idol`, `costume-dark`) also need a 全景 / 大全景 with negative
 space in every segment.
 For `sweet-romance` and `costume-idol`, shoot the 甜宠 set pieces (心动对视, 回眸, 共伞…) as
@@ -377,8 +395,10 @@ must be connected: pictures, then clips, then voices.
    spotting notes for the score.
 6. After each clip: `fs set <shot dataset> [--add-images <keyframes…>] --set primaryVideo=<clip>`;
    bind asset ids.
-7. **Score the edited episode.** When every clip is in, edit them into a locked cut (picture lock),
-   `wji asset upload` it and bind `wji.assets.edit`. `prompt score E0N [--fps 24]` prints the planned
+7. **Score the edited episode.** When every clip is in, cut them to picture lock as in
+   `reference/edit-tools.md`: FilmCraft (`filmcraft-cli`), timeline kept at `edits/E0N.fcproj`,
+   or ffmpeg when that CLI is not installed. `wji asset upload` the export and bind
+   `wji.assets.edit`. `prompt score E0N [--fps 24]` prints the planned
    spotting sheet — SMPTE rows for the scene beds, every line (where the music ducks), each
    segment's music cues at episode time, the `post` sound design, the sfx already in the clips, the
    hook / turns / 爽点 / cliff (起承转合) and the ≤120 s Seed Audio sections. Conform it to the edit's
@@ -386,7 +406,8 @@ must be connected: pictures, then clips, then voices.
    `reference/bgm-sfx.md`, loudness and terms in `reference/film-mixing-guideline.md`): BGM, BGE /
    ambience and sound design only — never what the clips already carry. One `canvas add audio -n
    "E0N score K" -m seed-audio-1.0 --prompt-file …` per section (~0.15 credits a minute — quote
-   first), mix the sections under the edit at their offsets, upload and bind `wji.assets.mix`.
+   first), lay the sections under the edit with SoundCraft, or ffmpeg when `soundcraft-cli` is
+   missing (`reference/edit-tools.md`), upload and bind `wji.assets.mix`.
    `manifest --ep N` must come back clean before you call the episode done. (`audioTrack: bed` /
    `full` instead make a Seed Audio node per segment from `prompt audio E0N-01` and put the music in
    the clip.)