@remixmate/cli 0.13.0 → 0.13.2

package/dist/manifest.json

@@ -1,7 +1,7 @@
 {
   "schemaVersion": 1,
-  "version": "0.13.0",
-  "generatedAt": "2026-10-25T02:08:12.079Z",
+  "version": "0.13.2",
+  "generatedAt": "2026-10-25T07:41:14.830Z",
   "skills": [
     {
       "id": "doc-parse",
@@ -602,6 +602,15 @@
             "type": "string",
             "description": "Media fill mode for templates declaring capabilities.mediaBinding.options.fit; one of that option's enum in the template's customPayloadSchema (its descriptions explain each value). Beats the preset's own fit. **Must** pass when the user picked it: `显示方式: <名称>（<id>）` → pass `<id>`; `auto` keeps the template / preset default."
           },
+          "narration": {
+            "type": "string",
+            "enum": [
+              "auto",
+              "on",
+              "off"
+            ],
+            "description": "Narration on / off, overriding the template's needsNarration (auto = the template decides). Templates where narration is optional (image sequences, galton-board) default to none — **must** pass `on` when the replicate card says `旁白: <名称>（narrated）`, otherwise the narration you write is dropped without an error and the video comes back silent. `off` drops narration from a narrated template."
+          },
           "background_asset": {
             "type": "string",
             "description": "URL of the user's background image or video (the replicate card's 背景素材 / 背景图片 asset line → its URL). Added as an existing asset and referenced from every scene's customPayload.background (assetRef for an image, videoAssetRef for a video; .mp4/.mov/.webm/.m4v count as video); the background preset becomes image / video, so background_preset can be omitted. **Must** pass when the user gave a background file — without it a preset of image / video renders the template's plain default. Requires template_id; rejected when the template has no background of that kind."
@@ -616,6 +625,7 @@
           "motion": {
             "type": "string",
             "enum": [
+              "auto",
               "static",
               "kenburns-in",
               "kenburns-out",
@@ -623,7 +633,7 @@
               "pan-right",
               "float"
             ],
-            "description": "Camera motion for every scene (scenes[].animationHints.motion), replacing the skeleton's alternating push-in / pull-out. **Must** pass when the user picked one: `运镜效果: 缓慢推近（kenburns-in）` → `kenburns-in`."
+            "description": "Camera motion for every scene (scenes[].animationHints.motion), replacing the skeleton's alternating push-in / pull-out. **Must** pass when the user picked one: `运镜效果: 缓慢推近（kenburns-in）` → `kenburns-in`; `运镜效果: 自动搭配（auto）` → `auto` (the template varies the motion scene by scene; do not write animationHints.motion yourself then)."
           },
           "ending_payload": {
             "type": "array",
@@ -737,6 +747,7 @@
           "background_preset",
           "sequence_preset",
           "sequence_fit",
+          "narration",
           "background_asset",
           "payload",
           "ending_payload",

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@remixmate/cli",
-  "version": "0.13.0",
+  "version": "0.13.2",
   "description": "AI media generation skills for Claude Code / Codex — 12 skills covering image, video, voice, digital human, web screenshot, web recording, script, template registry, rendering, Jianying export, and video deconstruction.",
   "type": "module",
   "main": "dist/index.js",

package/skills/gen-script/SKILL.md

@@ -484,11 +484,12 @@ python3 <SkillDir>/scripts/gen_script.py \
 | `--background-preset` | Scene background preset, one of the template's `customPayloadSchema.background.preset` values. Written to every scene's `customPayload.background.preset` (other keys of the template's default background are kept). Pass it whenever the user picked a background, e.g. `背景样式: <名称>（<id>）` → `<id>`. `image` / `video` need the user's file: pass `--background-asset` (which sets the preset itself). Requires `--template-id`; unsupported values are rejected. | template default |
 | `--sequence-preset` | Style preset for templates declaring `capabilities.mediaBinding.options.preset`; one of that option's enum in the template's `customPayloadSchema` (`none` keeps the template defaults). Written next to the media list; the preset sets transitions, motion, text layout and pacing, and the scene duration follows. **Must** pass when the user picked a style preset on the replicate card (pass the id in parentheses, e.g. `风格预设: <名称>（<id>）` → `<id>`); otherwise the choice is silently dropped. Rejected on templates without that option. |
 | `--sequence-fit` | Fill mode for templates declaring `capabilities.mediaBinding.options.fit`; one of that option's enum (`auto` keeps the template / preset default). Beats the preset's own fit. **Must** pass when the user picked it on the replicate card (`显示方式: <名称>（<id>）` → `<id>`). |
+| `--narration` | `auto` (the template's `capabilities.needsNarration` decides), `on` or `off`. Templates where narration is optional default to none; **must** pass `on` when the user picked narration on the replicate card (`旁白: <名称>（narrated）`), otherwise the narration written later is dropped without an error. `on` adds the narration slot, a `gen-voice` asset and subtitles; on `fit-images` templates that declare `customPayload.minDurationSec` it also writes the media's natural length there, so a short narration does not compress the media (a longer one stretches the scene). |
 | `--background-asset` | URL of the user's background image or video (`.mp4/.mov/.webm/.m4v` → video). Added as existing asset `asset-background`; every scene's `customPayload.background` gets `preset: image` + `assetRef`, or `preset: video` + `videoAssetRef`. Conflicts with a different `--background-preset`; rejected when the template's background enum lacks that kind. | none |
 | `--payload KEY=VALUE` | Template-specific card choice, repeatable. `KEY` is a dotted path in `customPayloadSchema` (or `capabilities.frameLevelKeys`), e.g. `<field>=<value>`, `<object>.<field>=<value>`. Value is converted to the declared type (true/false, numbers) and checked against enum / range; undeclared keys are rejected. Written to every scene with sibling keys kept, applied before fit-caption duration estimation. | template default |
 | `--ending-payload KEY=VALUE` | Same as `--payload`, but written only to the ending scene, e.g. `<object>.<field>=<closing line>`. On countdown templates it also appends an ending scene after #1 (it carries the #1 index; fill its narration and the item field `capabilities.countdown.itemProp` names). Errors when the script has no ending scene. | none |
 | `--avatar-id` | Digital-human avatar id (`dh_xxx`) for templates declaring `capabilities.digitalHuman`. Adds asset `asset-avatar` (`type: avatar`, `source: gen-digital-human`, `payload.avatarId`) and sets `visuals.avatar.assetRef` on every scene whose slot reads it — one continuous presenter driven by the narration. Placement: template default, or `--payload digitalHuman.<field>=…` (a photo-only presenter is `--payload digitalHuman.posterUrl=<url>` without this flag). | none |
-| `--motion` | Camera motion for every scene (`animationHints.motion`): `static` / `kenburns-in` / `kenburns-out` / `pan-left` / `pan-right` / `float`. | skeleton alternates in / out |
+| `--motion` | Camera motion for every scene (`animationHints.motion`): `static` / `kenburns-in` / `kenburns-out` / `pan-left` / `pan-right` / `float`. `auto` removes the motion from every scene so the template picks (image-slide rotates push / pan / pull / pan); nothing is locked then. | skeleton alternates in / out |
 | `--font-id` | Font library key (`font.uniq_id`), written to `global.font.fontId`; ab-render resolves and installs the real font at render time. Pass it whenever the user picked a font — the replicate card shows it as `名字（id）`, e.g. `字体: <名字>（<uniqId>）`; pass the id in the parentheses. Never pass a CSS family or the display name: an unknown key falls back to the template font with no error. | template font |
 | `--font-name` | Display name of the picked font → `global.font.fontName` (display only, never used for rendering). Pass it with `--font-id`. | — |
 | `--list-fonts` | Print the live font library (`/font/page`: id, name, lang, category, license) and exit; no script is generated and `--topic` is not needed. The ids it prints are the only valid `--font-id` values — use it when the user asks which fonts exist or names a font without its id. | — |

package/skills/gen-script/scripts/gen_script.py

@@ -582,6 +582,10 @@ def _template_primary_visual_type(template_config: dict | None) -> str:
     return "none"
 
 
+
+NARRATION_MODES = ("auto", "on", "off")
+
+
 def _template_needs_narration(template_config: dict | None) -> bool:
 
 
@@ -1382,10 +1386,6 @@ def _build_carousel_caption_dsl(
     _write_payload(custom_payload, payload_writes or [])
 
 
-    narration_text = _localize_label(output_language, "narration_opening",
-                                     label="", topic=_clip(topic, _PLACEHOLDER_TOPIC_CHARS), duration=duration)
-
-
 
 
 
@@ -1403,6 +1403,33 @@ def _build_carousel_caption_dsl(
             )
 
 
+
+
+    narration_text = _localize_label(
+        output_language, "narration_opening", label="", topic=_clip(topic, _PLACEHOLDER_TOPIC_CHARS),
+        duration=effective_duration,
+        budget=_budget_suffix(effective_duration, output_language, narration_speed, None),
+    )
+    narration_locks: list[str] = []
+    if narration_enabled and duration_strategy == "fit-images" and _schema_at(template_config, ["minDurationSec"]):
+
+
+
+        custom_payload["minDurationSec"] = effective_duration
+        narration_locks.append("customPayload.minDurationSec")
+
+
+
+
+        pacing_path = container_parts + ["pacing"]
+        pacing_schema = _schema_at(template_config, pacing_path)
+        explicit_pacing = any(parts == pacing_path for parts, _ in (payload_writes or []))
+        if (container_parts and not video_seconds and "fit-scene" in (pacing_schema.get("enum") or [])
+                and not explicit_pacing):
+            container["pacing"] = "fit-scene"
+            narration_locks.append("customPayload." + ".".join(container_parts + ["pacing"]))
+
+
     text_layers = [
         {"role": "headline", "content": headline, "animation": "slide-up"},
     ]
@@ -1459,6 +1486,8 @@ def _build_carousel_caption_dsl(
         "scenes": [scene],
         "transitions": {"default": "fade", "duration": 0.5},
     }
+    for path in narration_locks:
+        _lock_scene_path(dsl, path)
 
     return dsl
 
@@ -1844,18 +1873,23 @@ def _apply_scene_payload(
 
     writes = _parse_payload_specs(template_config, payload_specs)
     if motion:
-        allowed = ("static", "kenburns-in", "kenburns-out", "pan-left", "pan-right", "float")
+        allowed = ("auto", "static", "kenburns-in", "kenburns-out", "pan-left", "pan-right", "float")
         if motion not in allowed:
             raise ValueError(f"--motion {motion!r}: choose one of: {', '.join(allowed)}")
 
     for scene in dsl.get("scenes", []):
         _write_payload(scene.setdefault("customPayload", {}), writes)
-        if motion:
-            hints = scene.get("animationHints") if isinstance(scene.get("animationHints"), dict) else {}
-            scene["animationHints"] = {**hints, "motion": motion}
+        hints = scene.get("animationHints") if isinstance(scene.get("animationHints"), dict) else None
+        if motion == "auto":
+
+
+            if hints is not None:
+                hints.pop("motion", None)
+        elif motion:
+            scene["animationHints"] = {**(hints or {}), "motion": motion}
     for parts, _ in writes:
         _lock_scene_path(dsl, "customPayload." + ".".join(parts))
-    if motion:
+    if motion and motion != "auto":
         _lock_scene_path(dsl, "animationHints.motion")
 
 
@@ -2430,6 +2464,7 @@ def build_dsl(
         sequence_fit: str | None = None,
         payload_writes: PayloadWrites | None = None,
         countdown_ending: bool = False,
+        narration: str = "auto",
 ) -> dict:
 
 
@@ -2461,6 +2496,12 @@ def build_dsl(
 
     template_config = _load_template_config(template_id) if template_id else None
     contract = _resolve_contract(template_config)
+
+
+    if narration not in NARRATION_MODES:
+        raise ValueError(f"--narration must be one of {', '.join(NARRATION_MODES)}, got {narration!r}")
+    if narration != "auto":
+        contract["needs_narration"] = narration == "on"
     countdown = contract["countdown"]
     if countdown is None and item_count is not None:
         raise ValueError(
@@ -2751,6 +2792,13 @@ Examples:
              "the template defaults. When the user picked a style preset in the replicate card "
              "(`<名称>（<id>）`), pass that id.",
     )
+    parser.add_argument(
+        "--narration", default="auto", choices=NARRATION_MODES,
+        help="Narration on / off, overriding the template's capabilities.needsNarration (auto = the template "
+             "decides). Templates where narration is optional default to none; pass `on` when the user picked "
+             "narration in the replicate card (`旁白: <名称>（narrated）`), so the skeleton gets narration slots, "
+             "a gen-voice asset and subtitles. `off` drops narration from a narrated template.",
+    )
     parser.add_argument(
         "--sequence-fit", default=None,
         help="Media fill mode for templates declaring capabilities.mediaBinding.options.fit (one of that "
@@ -2774,7 +2822,8 @@ Examples:
     parser.add_argument(
         "--motion", default=None,
         help="Camera motion for every scene (animationHints.motion): static / kenburns-in / kenburns-out / "
-             "pan-left / pan-right / float. Replaces the skeleton's alternating default.",
+             "pan-left / pan-right / float. Replaces the skeleton's alternating default. `auto` removes the "
+             "motion from every scene so the template picks (image-slide rotates push / pan / pull / pan).",
     )
     parser.add_argument("--allow-digital-human", action="store_true", help="Allow digital-human assets")
     parser.add_argument("--avatar-id", default=None, help="Digital-human avatar id (dh_xxx) for templates declaring capabilities.digitalHuman: adds one gen-digital-human asset referenced by every scene's visuals.avatar.assetRef (one continuous presenter driven by the narration). Placement stays the template default unless set with --payload digitalHuman.<field>=...")
@@ -3084,6 +3133,7 @@ def main():
             sequence_fit=args.sequence_fit,
             payload_writes=payload_writes,
             countdown_ending=bool(args.ending_payload),
+            narration=args.narration,
         )
     except ValueError as exc:
 

package/skills/gen-script/skill.json

@@ -64,6 +64,11 @@
         "type": "string",
         "description": "Media fill mode for templates declaring capabilities.mediaBinding.options.fit; one of that option's enum in the template's customPayloadSchema (its descriptions explain each value). Beats the preset's own fit. **Must** pass when the user picked it: `显示方式: <名称>（<id>）` → pass `<id>`; `auto` keeps the template / preset default."
       },
+      "narration": {
+        "type": "string",
+        "enum": ["auto", "on", "off"],
+        "description": "Narration on / off, overriding the template's needsNarration (auto = the template decides). Templates where narration is optional (image sequences, galton-board) default to none — **must** pass `on` when the replicate card says `旁白: <名称>（narrated）`, otherwise the narration you write is dropped without an error and the video comes back silent. `off` drops narration from a narrated template."
+      },
       "background_asset": {
         "type": "string",
         "description": "URL of the user's background image or video (the replicate card's 背景素材 / 背景图片 asset line → its URL). Added as an existing asset and referenced from every scene's customPayload.background (assetRef for an image, videoAssetRef for a video; .mp4/.mov/.webm/.m4v count as video); the background preset becomes image / video, so background_preset can be omitted. **Must** pass when the user gave a background file — without it a preset of image / video renders the template's plain default. Requires template_id; rejected when the template has no background of that kind."
@@ -78,6 +83,7 @@
       "motion": {
         "type": "string",
         "enum": [
+          "auto",
           "static",
           "kenburns-in",
           "kenburns-out",
@@ -85,7 +91,7 @@
           "pan-right",
           "float"
         ],
-        "description": "Camera motion for every scene (scenes[].animationHints.motion), replacing the skeleton's alternating push-in / pull-out. **Must** pass when the user picked one: `运镜效果: 缓慢推近（kenburns-in）` → `kenburns-in`."
+        "description": "Camera motion for every scene (scenes[].animationHints.motion), replacing the skeleton's alternating push-in / pull-out. **Must** pass when the user picked one: `运镜效果: 缓慢推近（kenburns-in）` → `kenburns-in`; `运镜效果: 自动搭配（auto）` → `auto` (the template varies the motion scene by scene; do not write animationHints.motion yourself then)."
       },
       "ending_payload": {
         "type": "array",
@@ -199,6 +205,7 @@
       "background_preset",
       "sequence_preset",
       "sequence_fit",
+      "narration",
       "background_asset",
       "payload",
       "ending_payload",