@remixmate/cli 0.1.1 → 0.9.1

package/dist/runner.d.ts

@@ -7,7 +7,7 @@
  */
 import type { ParsedArgs } from './argv.js';
 export interface RunOptions {
-    /** Override the skills directory (defaults to ab-skill/skills). */
+    /** Override the skills directory (defaults to the package's bundled skills/). */
     baseDir?: string;
     /** Raw argv slice after the skill name (passed verbatim for python; parsed for handlers). */
     rawArgs: string[];

package/dist/skill-schema.d.ts

@@ -21,12 +21,30 @@ export type SkillEntry = {
     type: 'builtin';
     handler: string;
 };
+/**
+ * Skill 类别 — 区分"创作期 / 消费期 / 元数据型" skill。
+ *
+ * - "authoring": 创建 / 编辑模板期间使用，例如 template-registry / gen-script /
+ *                render-video（用于 try_render） / prepare-video-assets。
+ * - "consuming": 消费已有模板生成媒体，例如 export-jianying / video-parser /
+ *                web-capture（与"创建模板"无关，是消费侧场景）。
+ * - "asset":     生成单一类型素材的原子 skill：gen-image / gen-voice /
+ *                gen-video / gen-digital-human。authoring 也会用，但通过 stub
+ *                方式占位；consuming 场景下是真调用。
+ * - "meta":      元 skill（如未来加入 doc-only 的 template-creator）。
+ *
+ * 下游消费方（如 ab-template-studio）按 category 自动挑选白名单，避免每加一个
+ * skill 就要改两侧 hardcode。
+ */
+export type SkillCategory = 'authoring' | 'consuming' | 'asset' | 'meta';
+export declare const CATEGORY_VALUES: readonly ["authoring", "consuming", "asset", "meta"];
 export interface RawSkillJson {
     name: string;
     toolName: string;
     description: string;
     title?: string;
     tier?: string;
+    category?: SkillCategory;
     parameters?: Record<string, unknown>;
     scriptPath?: string;
     entry?: SkillEntry;

package/dist/skill-schema.js

@@ -11,6 +11,7 @@
  * human-readable problems (empty = OK). The runtime loader skips + warns;
  * the build treats them as fatal.
  */
+export const CATEGORY_VALUES = ['authoring', 'consuming', 'asset', 'meta'];
 export const TIER_VALUES = ['atomic', 'orchestration', 'tool'];
 export const REQUIRED_SKILL_JSON_FIELDS = ['name', 'tier', 'title', 'description'];
 /**
@@ -39,6 +40,9 @@ export function validateSkillJson(raw, skillId) {
     if (raw.tier != null && !TIER_VALUES.includes(raw.tier)) {
         errors.push(`skill.json.tier must be one of ${TIER_VALUES.join(' | ')}, got '${raw.tier}'`);
     }
+    if (raw.category != null && !CATEGORY_VALUES.includes(raw.category)) {
+        errors.push(`skill.json.category must be one of ${CATEGORY_VALUES.join(' | ')}, got '${raw.category}'`);
+    }
     if (raw.name && raw.name !== skillId) {
         errors.push(`skill.json.name='${raw.name}' does not match directory name '${skillId}'`);
     }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@remixmate/cli",
-  "version": "0.1.1",
+  "version": "0.9.1",
   "description": "AI media generation skills for Claude Code / Codex — 11 skills covering image, video, voice, digital human, web capture, script, template registry, rendering, Jianying export, and video deconstruction.",
   "type": "module",
   "main": "dist/index.js",

package/skills/export-jianying/skill.json

@@ -2,6 +2,7 @@
   "name": "export-jianying",
   "toolName": "export_jianying",
   "tier": "orchestration",
+  "category": "consuming",
   "title": "Jianying (CapCut) Draft Export",
   "description": "Jianying (CapCut) draft export: package asset URLs into a draft ZIP that Jianying can import. Supports automatic conversion from a RenderPlan.",
   "envVars": ["PRIV_TOKEN", "MM_API_BASE_URL", "AGENT_NAME"],

package/skills/gen-digital-human/SKILL.md

@@ -21,7 +21,7 @@ triggers:
 
 Wraps ab-api's digital-human endpoints (the same ones the web studio uses), authenticated with the **Tianyan privateToken**. Two providers: **Jimeng** (image-driven) and **HiFly** (video-driven). Generation is async — the handler submits the task and polls `/digital-human/video/status` until completion.
 
-> This skill was migrated from a Python script to an ab-skill CLI HTTP handler (`entry.type: http`). The agent invocation is unchanged (same tool name `gen_digital_human`, same params as in `skill.json`); local repro goes through `ab-skill gen-digital-human ...`.
+> This skill was migrated from a Python script to an remixmate CLI HTTP handler (`entry.type: http`). The agent invocation is unchanged (same tool name `gen_digital_human`, same params as in `skill.json`); local repro goes through `remixmate gen-digital-human ...`.
 
 ## Capabilities
 
@@ -54,24 +54,24 @@ By default only the URL is printed; generation is async and the handler polls un
 ### 1. List avatars
 
 ```bash
-ab-skill gen-digital-human --list-avatars
+remixmate gen-digital-human --list-avatars
 ```
 
 Optional filters:
 
 ```bash
 # Filter by source (jimeng = image-driven, hifly = video-driven)
-ab-skill gen-digital-human --list-avatars --source jimeng
-ab-skill gen-digital-human --list-avatars --source hifly
+remixmate gen-digital-human --list-avatars --source jimeng
+remixmate gen-digital-human --list-avatars --source hifly
 
 # Filter by gender
-ab-skill gen-digital-human --list-avatars --gender female
+remixmate gen-digital-human --list-avatars --gender female
 
 # Show your custom avatars
-ab-skill gen-digital-human --list-avatars --mine
+remixmate gen-digital-human --list-avatars --mine
 
 # Filter by name
-ab-skill gen-digital-human --list-avatars --name "alice"
+remixmate gen-digital-human --list-avatars --name "alice"
 ```
 
 > Custom avatars require `--mine`. When generating a video the handler auto-falls back from the public list to `mine: true` if needed.
@@ -81,7 +81,7 @@ ab-skill gen-digital-human --list-avatars --name "alice"
 #### TTS mode
 
 ```bash
-ab-skill gen-digital-human \
+remixmate gen-digital-human \
   --avatar-id 7 \
   --text "Hi everyone, welcome to the live stream — today I'll introduce a new product." \
   --voice-id "male-qn-qingse"
@@ -90,7 +90,7 @@ ab-skill gen-digital-human \
 #### TTS + custom voice + aspect ratio
 
 ```bash
-ab-skill gen-digital-human \
+remixmate gen-digital-human \
   --avatar-id 7 \
   --text "Hi class, here's our brand-new course." \
   --voice-id "female-shaonv" \
@@ -100,7 +100,7 @@ ab-skill gen-digital-human \
 #### Audio-driven mode
 
 ```bash
-ab-skill gen-digital-human \
+remixmate gen-digital-human \
   --avatar-id 39 \
   --source hifly \
   --audio-url "https://example.com/voice.mp3" \
@@ -110,7 +110,7 @@ ab-skill gen-digital-human \
 ### 3. Check job status
 
 ```bash
-ab-skill gen-digital-human --check-status --generation-id 123
+remixmate gen-digital-human --check-status --generation-id 123
 ```
 
 ## Common CLI flags

package/skills/gen-digital-human/skill.json

@@ -2,6 +2,7 @@
   "name": "gen-digital-human",
   "toolName": "gen_digital_human",
   "tier": "atomic",
+  "category": "asset",
   "title": "Digital-Human Talking-Head",
   "description": "Digital-human video: list available avatars; produce a talking-head video from text via TTS, or drive an avatar from an existing audio URL.",
   "envVars": ["PRIV_TOKEN", "MM_API_BASE_URL", "AGENT_NAME"],

package/skills/gen-digital-human/version.json

@@ -2,6 +2,6 @@
   "skillName": "gen-digital-human",
   "repoName": "agent-skill-media-maker",
   "skillId": "336",
-  "version": "V7",
+  "version": "V8",
   "skillDescription": "数字人口播视频技能，支持查询形象、TTS 口播、音频驱动口播和查询生成状态（调用 ab-api 数字人接口，即梦 / 飞影）。\n\n当用户提到以下任何需求时，立即使用本 skill：\n- 数字人、数字人视频、数字人口播、生成数字人视频\n- 用户提供已有音频 URL（MP3 等）、用外链音频驱动数字人口型\n- AI 主播、虚拟主播、虚拟人物视频、口播视频\n- 让数字人说话、让虚拟人说一段话、让 AI 人物播报\n- 查看/列出数字人形象、有哪些数字人可以用\n- 使用即梦/飞影数字人\n\n即使用户没有明确说「使用 AI」，只要他们想要让一个虚拟人物朗读/播报一段文字并生成视频，也要使用本 skill。"
 }

package/skills/gen-image/SKILL.md

@@ -51,7 +51,7 @@ No skill-local env file — the executing process inherits the system environmen
 
 ## Operations
 
-> This skill was migrated from a Python script to an ab-skill CLI HTTP handler (`entry.type: http`). The agent invocation is unchanged (same tool name `gen_image`, same params as in `skill.json`); local repro goes through `ab-skill gen-image ...`.
+> This skill was migrated from a Python script to an remixmate CLI HTTP handler (`entry.type: http`). The agent invocation is unchanged (same tool name `gen_image`, same params as in `skill.json`); local repro goes through `remixmate gen-image ...`.
 
 1. **Prompt**: be specific about subject, style, lighting, composition. Either English or Chinese works.
 2. By default only the URL is printed (good for showing to the user); the legacy local-download flag has been dropped — image URLs are persisted in the cloud.
@@ -59,13 +59,13 @@ No skill-local env file — the executing process inherits the system environmen
 ### Text-to-image
 
 ```bash
-ab-skill gen-image \
+remixmate gen-image \
   --prompt "<image description>" \
   --size "9:16"
 ```
 
 ```bash
-ab-skill gen-image \
+remixmate gen-image \
   --prompt "<image description>" \
   --model gemini-3-pro-image \
   --size "16:9" \
@@ -81,14 +81,14 @@ Reference images accept local file paths, HTTPS URLs, or data URIs. Pass `--refe
 
 ```bash
 # URL reference
-ab-skill gen-image \
+remixmate gen-image \
   --prompt "Convert this photo to an oil-painting style" \
   --reference "https://example.com/photo.jpg"
 ```
 
 ```bash
 # Local-file reference + reference strength
-ab-skill gen-image \
+remixmate gen-image \
   --prompt "Match the style of this reference" \
   --reference ./ref.png \
   --image-strength 0.6
@@ -96,7 +96,7 @@ ab-skill gen-image \
 
 ```bash
 # Multiple references
-ab-skill gen-image \
+remixmate gen-image \
   --prompt "Blend these styles" \
   --reference ./a.png \
   --reference ./b.png

package/skills/gen-image/skill.json

@@ -2,6 +2,7 @@
   "name": "gen-image",
   "toolName": "gen_image",
   "tier": "atomic",
+  "category": "asset",
   "title": "AI Image Generation",
   "description": "AI image generation: produce an image from a text prompt. Supports Seedream and Gemini models, plus image-to-image with reference images.",
   "envVars": ["PRIV_TOKEN", "MM_API_BASE_URL", "AGENT_NAME", "MM_IMAGE_MODEL"],

package/skills/gen-image/version.json

@@ -2,6 +2,6 @@
   "skillName": "gen-image",
   "repoName": "agent-skill-media-maker",
   "skillId": "337",
-  "version": "V8",
+  "version": "V9",
   "skillDescription": "AI 生图技能，根据文字描述生成图片，也支持参考图进行图生图（调用 ab-api /model/genImg，支持 Seedream 与 Gemini）。\n\n当用户提到以下任何需求时，立即使用本 skill：\n- AI 生图、AI 画图、文生图、文字生成图片、生成图像、帮我画、生成一张图\n- 图生图、以图生图、参考图、风格迁移、图片变体\n- 使用 doubao / 豆包 / seedream、Gemini 等生成图片\n- 用户提供图片提示词并希望生成图片\n\n即使用户没有明确说「使用 AI」，只要他们想要根据描述生成图片，也要使用本 skill。"
 }

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

@@ -322,9 +322,9 @@ def generate_image_prompt(
 def _load_template_config(template_id: str) -> dict | None:
     """Load template definition for the given template ID.
 
-    解析顺序（与 match_template / template_binder 保持一致）：
-      1. 通过 registry_loader 从 monorepo template-library/packages/metadata/registry.json
-         （或 VIDEO_TEMPLATE_REGISTRY[_URL] 环境变量）加载，并按 templateId 匹配。
+    解析顺序：
+      1. 通过 registry_loader 从 ab-api（单一数据源，URL 取
+         VIDEO_TEMPLATE_REGISTRY_URL 或按 MM_API_BASE_URL 推导）加载，按 templateId 匹配。
       2. 兜底：旧路径 template-registry/video_dsl/templates/<id>/template.json
          （仅在仓库还残留旧目录时使用）。
     返回完整 template dict（含 supportedAspectRatios / assetRequirements / slotMapping 等），
@@ -384,6 +384,29 @@ def _template_primary_visual_type(template_config: dict | None) -> str:
     return "none"
 
 
+def _template_needs_narration(template_config: dict | None) -> bool:
+    """模板是否需要旁白（TTS）。
+
+    默认 True —— 保持历史行为：每个场景挂一个 gen-voice 素材 + 写入
+    ``audio.narration`` 骨架，由 agent 后续填真实文案。
+
+    模板可通过 ``capabilities.needsNarration=false`` 声明"纯视觉 / 无旁白"
+    （如单图 Ken Burns、BGM-only 展示）。此时 gen_script：
+      - 不生成 gen-voice 素材；
+      - 场景不写 ``audio.narration``。
+    于是骨架的 narrationSceneCount=0，ab-agent 的 prepare_video_assets 预校验
+    会直接放行，不再强制用户为每个场景填旁白（与 spotlight-card 等无配音模板
+    走的是同一条 pass-through 路径）。
+
+    兼容历史脏数据：部分模板的 ``capabilities`` 是空列表 ``[]``（而非 dict），
+    统一按"未声明"处理 → 返回 True。仅当显式为 JSON ``false`` 时才关闭旁白。
+    """
+    caps = (template_config or {}).get("capabilities")
+    if not isinstance(caps, dict):
+        return True
+    return caps.get("needsNarration", True) is not False
+
+
 def _template_needs_image(template_config: dict | None) -> bool:
     """Check if a template requires image assets.
 
@@ -786,6 +809,9 @@ def build_dsl(
             )
     primary_visual_type = _template_primary_visual_type(template_config)
     has_visual = primary_visual_type in ("image", "video")
+    # 旁白是模板能力（capabilities.needsNarration）。声明为 false 的模板（纯视觉 /
+    # BGM-only，如单图 Ken Burns）不分配 gen-voice 素材、场景不写 audio.narration。
+    needs_narration = _template_needs_narration(template_config)
 
     # ── Resolve template-driven output language + voice ────────────────────
     # outputLanguage is owned by the template (template.json). Missing/invalid
@@ -815,8 +841,11 @@ def build_dsl(
             visual_asset_id = f"video-{scene_id}"
         else:
             visual_asset_id = f"img-{scene_id}"
-        narration_asset_id = f"narration-{scene_id}"
-        narration_text = estimate_narration_text(plan["label"], plan["purpose"], plan["duration"], topic, language=output_language)
+        narration_asset_id = f"narration-{scene_id}" if needs_narration else None
+        narration_text = (
+            estimate_narration_text(plan["label"], plan["purpose"], plan["duration"], topic, language=output_language)
+            if needs_narration else ""
+        )
 
         # 根据模板 assetRequirements 生成对应类型的视觉素材
         if primary_visual_type == "image":
@@ -881,15 +910,17 @@ def build_dsl(
         # narration text 不再在 audio asset 的 payload 里冗余存放——
         # 唯一来源是下游 scenes[].audio.narration.text，render_video 在
         # 调用 TTS skill 前会按 assetRef 回查 scene 文本注入。
-        assets.append({
-            "assetId": narration_asset_id,
-            "type": "audio",
-            "source": "gen-voice",
-            "status": "planned",
-            "payload": {
-                "voiceId": resolved_voice_id,
-            },
-        })
+        # 模板声明 needsNarration=false 时，完全不生成 gen-voice 素材。
+        if needs_narration:
+            assets.append({
+                "assetId": narration_asset_id,
+                "type": "audio",
+                "source": "gen-voice",
+                "status": "planned",
+                "payload": {
+                    "voiceId": resolved_voice_id,
+                },
+            })
 
         if has_visual:
             layout = "text-overlay" if plan["purpose"] in ("opening", "cta") else "full-visual"
@@ -916,33 +947,35 @@ def build_dsl(
                 if effective_subheadline:
                     text_layers.append({"role": "subheadline", "content": effective_subheadline, "animation": "fade-in"})
 
-            scenes.append({
+            scene = {
                 "id": scene_id,
                 "purpose": plan["purpose"],
                 "duration": plan["duration"],
                 "layout": layout,
                 "visuals": {"background": {"assetRef": visual_asset_id}},
-                "audio": {
+                "textLayers": text_layers,
+                "animationHints": {
+                    "entrance": "fade",
+                    "motion": "kenburns-in" if idx % 2 == 0 else "kenburns-out",
+                },
+            }
+            if needs_narration:
+                scene["audio"] = {
                     "narration": {
                         "text": narration_text,
                         "assetRef": narration_asset_id,
                         # 骨架标记：scenes[].audio.narration.text 同样需要被真实旁白替换
                         "needsFill": True,
                     },
-                },
-                "textLayers": text_layers,
-                "animationHints": {
-                    "entrance": "fade",
-                    "motion": "kenburns-in" if idx % 2 == 0 else "kenburns-out",
-                },
-            })
+                }
+            scenes.append(scene)
         else:
             # The "no-visual" branch labels each scene with a layout hint.
             # Historical fallback was hardcoded "html-slide" — when no template
             # was specified, every audio-only scene defaulted to html-slide
             # styling. We replace the hardcoded fallback chain with:
             #   1. template.capabilities.defaultLayout   ⇐ the template tells
-            #      ab-skill what layout name it wants on its no-visual scenes
+            #      remixmate what layout name it wants on its no-visual scenes
             #   2. template_id                            ⇐ legacy: pass id as
             #      layout name (existing behavior when no capability declared)
             #   3. "html-slide"                           ⇐ absolute fallback
@@ -966,22 +999,24 @@ def build_dsl(
 
             custom_payload = _build_custom_payload(plan["purpose"], topic, plan["label"], idx, len(scene_plans), language=output_language)
 
-            scenes.append({
+            scene = {
                 "id": scene_id,
                 "purpose": plan["purpose"],
                 "duration": plan["duration"],
                 "layout": layout,
-                "audio": {
+                "textLayers": text_layers,
+                "customPayload": custom_payload,
+            }
+            if needs_narration:
+                scene["audio"] = {
                     "narration": {
                         "text": narration_text,
                         "assetRef": narration_asset_id,
                         # 骨架标记：scenes[].audio.narration.text 同样需要被真实旁白替换
                         "needsFill": True,
                     },
-                },
-                "textLayers": text_layers,
-                "customPayload": custom_payload,
-            })
+                }
+            scenes.append(scene)
 
     dsl = {
         "version": "v1alpha1",
@@ -1001,7 +1036,7 @@ def build_dsl(
             "aspectRatio": ratio,
             "resolution": resolution,
             "fps": 30,
-            "subtitle": {"enabled": True, "style": "bottom"},
+            "subtitle": {"enabled": needs_narration, "style": "bottom"},
             "narration": {"voiceId": resolved_voice_id, "speed": 1.0},
             "bgm": {"enabled": True, "volume": 0.12},
         },
@@ -1054,6 +1089,11 @@ Examples:
         default=None,
         help="On-screen subheadline / project name (e.g. 'Pixelle-Video'). Stored at meta.subheadline and pushed into textLayers[role=subheadline]. Independent from CC subtitles (global.subtitle).",
     )
+    parser.add_argument(
+        "--skip-asset-generation",
+        action="store_true",
+        help="Skill-creator / template-creator helper: when set, every produced AssetRef is marked as already generated with a stub URL — no gen-image / gen-voice / gen-video calls are needed. Implies --stub-image-url + --stub-video-url with default sentinels (https://placeholder.local/stub.png|.mp4) when those flags are absent, and additionally rewrites every gen-voice asset to source=existing + status=generated + a placeholder audio URL. Useful when a downstream agent only wants the DSL shape (e.g. to feed into try_render with all assets pre-stubbed).",
+    )
     parser.add_argument(
         "--stub-image-url",
         default=None,
@@ -1157,6 +1197,22 @@ Examples:
             print(f"⚠️  STUB_VIDEO_URL env var detected ({env_v}); using it as the video stub. Prefer passing --stub-video-url explicitly, or unset the env var.", file=sys.stderr)
         stub_video_url = env_v
 
+    # --skip-asset-generation 是为下游"只想要 DSL shape"的 agent 设计的
+    # 一键开关：等价于 --stub-image-url + --stub-video-url + 把 gen-voice 资产
+    # 也写成 source=existing + status=generated。当用户没显式提供 stub URL 时
+    # 用一组 sentinel 占位（https://placeholder.local/...），模板创作 / 调试场景
+    # 不会真的去 fetch 这些 URL。
+    if args.skip_asset_generation:
+        if not stub_image_url:
+            stub_image_url = "https://placeholder.local/stub.png"
+        if not stub_video_url:
+            stub_video_url = "https://placeholder.local/stub.mp4"
+        print(
+            "ℹ️  --skip-asset-generation: forcing all assets to source=existing/status=generated "
+            f"(image={stub_image_url}, video={stub_video_url}, audio=https://placeholder.local/stub.mp3)",
+            file=sys.stderr,
+        )
+
     dsl = build_dsl(
         topic=args.topic,
         platform=args.platform,
@@ -1177,6 +1233,24 @@ Examples:
         caption_lines=args.caption_lines,
     )
 
+    # Post-process for --skip-asset-generation: rewrite all gen-voice / gen-digital-human
+    # assets to be already-generated stubs. Image / video are already covered by the
+    # stub_image_url / stub_video_url params threaded through build_dsl above.
+    if args.skip_asset_generation:
+        _AUDIO_STUB = "https://placeholder.local/stub.mp3"
+        _VIDEO_STUB = stub_video_url  # same sentinel for digital-human placeholder
+        for asset in dsl.get("assets", []):
+            src = asset.get("source")
+            if src in ("gen-voice",):
+                asset["source"] = "existing"
+                asset["status"] = "generated"
+                asset["url"] = _AUDIO_STUB
+                # payload 留作 reference（renderer 不会再读它，因为 status=generated）
+            elif src in ("gen-digital-human",):
+                asset["source"] = "existing"
+                asset["status"] = "generated"
+                asset["url"] = _VIDEO_STUB
+
     errors = validate_dsl(dsl)
     if errors:
         print("❌ generated DSL failed validation:", file=sys.stderr)

package/skills/gen-script/skill.json

@@ -2,6 +2,7 @@
   "name": "gen-script",
   "toolName": "gen_script",
   "tier": "orchestration",
+  "category": "authoring",
   "title": "Video Script Generation",
   "description": "Video script generation: turn a topic into a structured Video DSL (JSON) that describes the full video — scene structure, asset requirements, and narrative flow.",
   "envVars": ["DEFAULT_IMAGE_MODEL", "DEFAULT_VIDEO_MODEL", "STUB_IMAGE_URL", "STUB_VIDEO_URL"],
@@ -34,7 +35,8 @@
         "description": "Bottom typewriter text lines for templates that support a caption/typewriter area (e.g. spotlight-card). Each element is one line of text. Supports **emphasis** syntax (rendered with accent color). **Must** pass when the user explicitly provides bullet-point text / bottom copy for the video."
       },
       "stub_image_url": { "type": "string", "description": "Test-mode image stub URL. Only pass when the user explicitly says things like 'just testing / don't actually generate / use a placeholder image / stub URL / save credits' AND provides a concrete URL. With this set, every image AssetRef in the produced DSL is written as source=existing, status=generated, url=<this URL> — no gen-image call. Do not pass otherwise; if the user expressed the intent without a URL, ask for one — do not invent one." },
-      "stub_video_url": { "type": "string", "description": "Test-mode video stub URL. Only pass when the user explicitly says things like 'just testing / don't actually generate the video / placeholder clip / save credits' AND provides a concrete URL. With this set, every video AssetRef in the produced DSL is written as source=existing, status=generated, url=<this URL> — no gen-video call. Do not pass otherwise; if the user expressed the intent without a URL, ask for one — do not invent one." }
+      "stub_video_url": { "type": "string", "description": "Test-mode video stub URL. Only pass when the user explicitly says things like 'just testing / don't actually generate the video / placeholder clip / save credits' AND provides a concrete URL. With this set, every video AssetRef in the produced DSL is written as source=existing, status=generated, url=<this URL> — no gen-video call. Do not pass otherwise; if the user expressed the intent without a URL, ask for one — do not invent one." },
+      "skip_asset_generation": { "type": "boolean", "description": "All-in-one switch for downstream agents (e.g. template-creator) that only want the DSL shape: every produced AssetRef is marked as already generated with placeholder URLs (image: https://placeholder.local/stub.png, video: stub.mp4, audio: stub.mp3). Implies the equivalent of --stub-image-url + --stub-video-url with sentinel defaults plus the same rewrite for gen-voice / gen-digital-human assets. Useful when the agent only needs to inspect DSL structure or feed it into try_render_local with all assets pre-stubbed." }
     },
     "required": ["topic"]
   }

package/skills/gen-script/version.json

@@ -2,6 +2,6 @@
   "skillName": "gen-script",
   "repoName": "agent-skill-media-maker",
   "skillId": "474",
-  "version": "V8",
+  "version": "V10",
   "skillDescription": "视频脚本生成技能，将用户主题转化为结构化 Video DSL（JSON），描述视频的完整结构、素材需求与叙事逻辑。\n\n当用户提到以下任何需求时，立即使用本 skill：\n- 帮我写视频脚本、生成视频脚本、视频策划、写分镜脚本\n- 做一个短视频、帮我规划视频内容、生成视频 DSL\n- 把主题转成视频结构、视频内容规划\n\n即使用户没有明确说「生成 DSL」，只要他们想要把一个主题变成视频内容结构，也要使用本 skill。"
 }

package/skills/gen-video/SKILL.md

@@ -19,7 +19,7 @@ triggers:
 
 Wraps ab-api's `POST /model/genVideo` (the same endpoint the web "Lingchuang AI Video" tool uses), authenticated with the **Tianyan privateToken**, routed through LiteLLM to **Seedance** or **Veo**. Generation is async — the handler submits the task and polls `/model/getVideoStatus` until completion.
 
-> This skill was migrated from a Python script to an ab-skill CLI HTTP handler (`entry.type: http`). The agent invocation is unchanged (same tool name `gen_video`, same params as in `skill.json`); local repro goes through `ab-skill gen-video ...`.
+> This skill was migrated from a Python script to an remixmate CLI HTTP handler (`entry.type: http`). The agent invocation is unchanged (same tool name `gen_video`, same params as in `skill.json`); local repro goes through `remixmate gen-video ...`.
 
 ## Models
 
@@ -72,7 +72,7 @@ No skill-local env file — the executing process inherits the system environmen
 ### Seedance text-to-video (default model)
 
 ```bash
-ab-skill gen-video \
+remixmate gen-video \
   --prompt "<video description>" \
   --duration 5 \
   --ratio "16:9"
@@ -83,7 +83,7 @@ ab-skill gen-video \
 Frame images accept local file paths, HTTPS URLs, or data URIs (local files are base64-encoded into a data URI).
 
 ```bash
-ab-skill gen-video \
+remixmate gen-video \
   --prompt "<transition description>" \
   --first-frame ./start.png \
   --last-frame ./end.png \
@@ -94,7 +94,7 @@ ab-skill gen-video \
 ### Veo 3.1 high-resolution
 
 ```bash
-ab-skill gen-video \
+remixmate gen-video \
   --model veo \
   --prompt "<video description>" \
   --duration 8 \
@@ -105,7 +105,7 @@ ab-skill gen-video \
 ### Veo 3.1 Fast for rapid iteration
 
 ```bash
-ab-skill gen-video \
+remixmate gen-video \
   --model veo-fast \
   --prompt "<video description>" \
   --duration 6
@@ -116,7 +116,7 @@ ab-skill gen-video \
 Pass `--reference` multiple times (Veo only, up to 3).
 
 ```bash
-ab-skill gen-video \
+remixmate gen-video \
   --model veo \
   --prompt "<video description>" \
   --reference ./ref1.png \

package/skills/gen-video/skill.json

@@ -2,6 +2,7 @@
   "name": "gen-video",
   "toolName": "gen_video",
   "tier": "atomic",
+  "category": "asset",
   "title": "AI Video Generation",
   "description": "AI video generation: produce a short video clip from a text prompt. Supports Seedance and Veo models, plus first/last frame and reference images.",
   "envVars": ["PRIV_TOKEN", "MM_API_BASE_URL", "AGENT_NAME", "MM_VIDEO_MODEL"],

package/skills/gen-video/version.json

@@ -2,6 +2,6 @@
   "skillName": "gen-video",
   "repoName": "agent-skill-media-maker",
   "skillId": "339",
-  "version": "V7",
+  "version": "V8",
   "skillDescription": "AI 生视频技能，根据文字描述生成素材视频（调用 ab-api /model/genVideo，支持 Seedance 与 Veo）。\n\n当用户提到以下任何需求时，立即使用本 skill：\n- AI 生视频、文生视频、文字生成视频、生成一段视频、AI 制作视频\n- 使用 doubao / 豆包 / seedance、Veo、Google 等生成视频\n- 用户提供视频提示词并希望生成视频\n- 图生视频、首帧生成视频、参考图生成视频\n\n即使用户没有明确说「使用 AI」，只要他们想要根据描述生成视频，也要使用本 skill。"
 }

package/skills/gen-voice/SKILL.md

@@ -34,7 +34,7 @@ There is no skill-local env file — the executing process inherits the system e
 
 ## Operations
 
-> This skill was migrated from a Python script to an ab-skill CLI HTTP handler (`entry.type: http`). The agent invocation is unchanged (same tool name `gen_voice`, same params as in `skill.json`); local repro goes through `ab-skill gen-voice ...`. The legacy `--download` flag has been removed — audio URLs are persisted in the cloud and play directly.
+> This skill was migrated from a Python script to an remixmate CLI HTTP handler (`entry.type: http`). The agent invocation is unchanged (same tool name `gen_voice`, same params as in `skill.json`); local repro goes through `remixmate gen-voice ...`. The legacy `--download` flag has been removed — audio URLs are persisted in the cloud and play directly.
 
 1. **Text**: confirm what to synthesize. Punctuation drives pauses (commas short, periods long).
 2. **Voice**: if the user prefers a specific voice, run `--list-voices` first and pick a matching id.
@@ -43,13 +43,13 @@ There is no skill-local env file — the executing process inherits the system e
 ### List available voices
 
 ```bash
-ab-skill gen-voice --list-voices
+remixmate gen-voice --list-voices
 ```
 
 To inspect the local language-tagged fallback catalog used by the voice resolver (no remote API call), add `--local`:
 
 ```bash
-ab-skill gen-voice --list-voices --local
+remixmate gen-voice --list-voices --local
 ```
 
 The local catalog prints one voice per line as `<voice-id>\t<lang>\t<display-name>`.
@@ -57,13 +57,13 @@ The local catalog prints one voice per line as `<voice-id>\t<lang>\t<display-nam
 ### Default synthesis (URL output)
 
 ```bash
-ab-skill gen-voice --text "<text-to-synthesize>"
+remixmate gen-voice --text "<text-to-synthesize>"
 ```
 
 ### With voice + speed
 
 ```bash
-ab-skill gen-voice \
+remixmate gen-voice \
   --text "<text-to-synthesize>" \
   --voice-id "female-shaonv" \
   --speed 1.2
@@ -72,7 +72,7 @@ ab-skill gen-voice \
 ### JSON output (with subtitle timestamps)
 
 ```bash
-ab-skill gen-voice --text "<text-to-synthesize>" --json-output
+remixmate gen-voice --text "<text-to-synthesize>" --json-output
 ```
 
 ## Common CLI flags

package/skills/gen-voice/skill.json

@@ -2,6 +2,7 @@
   "name": "gen-voice",
   "toolName": "gen_voice",
   "tier": "atomic",
+  "category": "asset",
   "title": "Text-to-Speech (Minimax)",
   "description": "Text-to-speech (TTS): synthesize narration audio from text via the Minimax TTS model. Returns the persisted audio URL — no download needed.",
   "envVars": ["PRIV_TOKEN", "MM_API_BASE_URL", "AGENT_NAME"],