@remixmate/cli 0.9.53 → 0.9.54

package/dist/manifest.json

@@ -1,7 +1,7 @@
 {
   "schemaVersion": 1,
-  "version": "0.9.53",
-  "generatedAt": "2026-09-28T02:19:31.947Z",
+  "version": "0.9.54",
+  "generatedAt": "2026-09-28T11:58:29.708Z",
   "skills": [
     {
       "id": "doc-parse",

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@remixmate/cli",
-  "version": "0.9.53",
+  "version": "0.9.54",
   "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/prepare-video-assets/version.json

@@ -2,6 +2,6 @@
   "skillName": "prepare-video-assets",
   "repoName": "agent-skill-media-maker",
   "skillId": "476",
-  "version": "V4",
+  "version": "V5",
   "skillDescription": "Asset-preparation skill: resolves and generates every asset (image / audio / video) referenced by a Video DSL, persists a RenderPlan to the database, and returns a `job_id` for the subsequent `render_video` call.\n\nUse this skill as soon as the user mentions any of these intents:\n- Generate / prepare video assets, resolve assets, render-ready\n- \"Make me a video about X\" (the agent calls gen_script → prepare_video_assets → render_video)\n- Regenerate one asset (image / audio) for a specific scene\n\nNext step: after the user confirms the resolved assets, call `render_video` with the `job_id` returned by this skill.\n\n⚠️ Stop-and-confirm gate: this skill runs only after the user has confirmed the script, and after it returns you must show the resolved assets and wait for the user's explicit confirmation. Never call `render_video` in the same turn."
 }

package/skills/render-video/scripts/render_video.py

@@ -196,6 +196,23 @@ def _narration_language_boost(narration: dict) -> Optional[str]:
     return raw.strip()
 
 
+def _narration_voice(narration: dict) -> Optional[str]:
+    """Read a narration block's voice id, or None when unset.
+
+    Same tolerance as _narration_speed. Exists because the voice used to be read
+    only from the audio asset's payload.voiceId — which gen_script bakes once
+    when it drafts the skeleton — so changing global.narration.voiceId (or a
+    scene's voice) after that never reached TTS: the video kept the original
+    voice while every other layer reported the new one (2026-09-28).
+    """
+    if not isinstance(narration, dict):
+        return None
+    raw = narration.get("voiceId")
+    if not isinstance(raw, str) or not raw.strip():
+        return None
+    return raw.strip()
+
+
 def extract_narration_lines(narration: dict) -> Optional[tuple[list[str], int, list[Optional[float]]]]:
     """If narration uses the structured {intro, items, outro} form, return
     (lines, intro_line_count, at_sec_list). `at_sec_list` is parallel to
@@ -783,6 +800,7 @@ def build_render_plan(dsl: dict, binding: dict) -> dict:
     global_narration = (dsl.get("global") or {}).get("narration") or {}
     global_speed = _narration_speed(global_narration)
     global_language_boost = _narration_language_boost(global_narration)
+    global_voice = _narration_voice(global_narration)
     for scene in dsl.get("scenes", []):
         narration = (scene.get("audio") or {}).get("narration") or {}
         ref = narration.get("assetRef")
@@ -790,6 +808,14 @@ def build_render_plan(dsl: dict, binding: dict) -> dict:
         if not asset:
             continue
         payload = asset.setdefault("payload", {})
+        # Voice follows the same precedence as speed: scene narration.voiceId >
+        # global.narration.voiceId. The asset's own payload.voiceId (written once
+        # by gen_script) is only the last resort for DSLs that declare neither —
+        # letting it win meant a voice change made after the script was drafted
+        # never reached TTS.
+        voice = _narration_voice(narration) or global_voice
+        if voice:
+            payload["voiceId"] = voice
         # Speech rate follows the same route as the narration text: single source
         # of truth on the DSL, copied down here because resolve_asset_audio only
         # ever sees the asset. Scene-level overrides global (mirrors the narration

package/skills/render-video/version.json

@@ -2,6 +2,6 @@
   "skillName": "render-video",
   "repoName": "agent-skill-media-maker",
   "skillId": "473",
-  "version": "V31",
+  "version": "V32",
   "skillDescription": "Final-render skill: loads a persisted RenderPlan by `job_id` and drives the Remotion engine to produce the final video.\n\nUse this skill as soon as the user mentions any of these intents (after assets are already prepared):\n- Render the video, composite the video, export the video\n- Turn the prepared assets into the final clip\n- Render with Remotion\n\nPrerequisite: assets must already be generated via `prepare_video_assets`. This skill never resolves or regenerates assets — pass it a `job_id` from a previous `prepare_video_assets` call.\n\n⚠️ Stop-and-confirm gate: never call this skill until the user has explicitly confirmed the assets prepared by `prepare_video_assets`. If those assets were prepared in the current turn and the user has not replied since, stop and ask instead of rendering."
 }

package/skills/template-registry/SKILL.md

@@ -169,6 +169,7 @@ Rules:
 - If the author explicitly provides a `highlightMap`, the system keeps it as-is.
 - The legacy flat `narration.text` form still works but the author then has to keep `highlightMap` in sync with the subtitle segmentation (error-prone, not recommended).
 - `narration.languageBoost` (optional, also accepted at `global.narration`) pins the TTS language hint (`auto` / `Chinese` / `English` / …). Leave it unset unless auto-detection gets it wrong — the service applies its own default, and a scene-level value overrides the global one, exactly like `speed`.
+- Voice: to change the narration voice, set `global.narration.voiceId` (whole video) or `narration.voiceId` on a scene (that scene only). Precedence when synthesizing is scene > global > the audio asset's `payload.voiceId`, which is only a fallback — editing the asset payload alone is not how you change a voice.
 
 ## Image sources
 

package/skills/template-registry/version.json

@@ -2,6 +2,6 @@
   "skillName": "template-registry",
   "repoName": "agent-skill-media-maker",
   "skillId": "475",
-  "version": "V24",
+  "version": "V25",
   "skillDescription": "Video-template registry skill. Stores every video-template definition and lists the available templates (templateId / name / aspect ratio / style tags).\n\nUse this skill as soon as the user mentions any of these intents:\n- View available templates / list every template\n\nNote: DSL→TemplateBinding is no longer a separate exposed step — once prepare_video_assets receives a template_id it builds the binding internally."
 }

package/skills/template-registry/video_dsl/schema/video-dsl-v1alpha1.json

@@ -247,6 +247,9 @@
                 },
                 "outro": { "type": "string", "description": "可选：收尾句（此期间无卡片高亮）" },
                 "assetRef": { "type": "string", "description": "旁白音频 assetId" },
+                "voiceId": { "type": "string", "description": "该场旁白音色，覆盖 global.narration.voiceId；留空跟随全局。合成时优先级：场景 > 全局 > 音频资产 payload.voiceId（仅作兜底）。" },
+                "voiceName": { "type": "string", "description": "voiceId 的展示名（仅展示用，不参与合成）" },
+                "speed": { "type": "number", "description": "该场旁白语速，覆盖 global.narration.speed；留空跟随全局。" },
                 "languageBoost": { "type": "string", "description": "TTS 语种提示（auto / Chinese / English …）。中英混排的旁白不给提示时，英文词容易被按中文音系读出来；覆盖 global.narration 的同名字段；留空则跟随全局/服务端默认。" }
               }
             },