@mindstudio-ai/agent 0.1.151 → 0.1.152

package/dist/index.d.ts

@@ -5597,6 +5597,10 @@ interface DecomposeImageStepInput {
     image: string;
     /** Optional guidance for which objects or groupings to separate */
     prompt?: string;
+    /** Decomposition quality tier: "standard" (default) or "advanced" */
+    type?: "standard" | "advanced";
+    /** Output resolution: "1k", "1.5k" or "2k" (default) */
+    resolution?: "1k" | "1.5k" | "2k";
     /** When true, assets are created but hidden from the user's gallery */
     intermediateAsset?: boolean;
 }
@@ -7475,7 +7479,7 @@ interface MeshyRigStepOutput {
 interface MeshyTextTo3dStepInput {
     /** Description of the 3D model to generate (max 600 characters) */
     prompt: string;
-    /** "standard" (default) or "lowpoly". Lowpoly ignores topology/target_polycount. */
+    /** "standard" (default) or "smart-topology", clean low-poly geometry (target_polycount 100-15000). "lowpoly", which Meshy retires on 2026-10-30, runs as smart-topology. */
     modelType?: string;
     /** "triangle" (default) or "quad" */
     topology?: string;
@@ -9512,8 +9516,10 @@ interface StepMethods {
      * Analyze a video using a video analysis model based on a text prompt.
      *
      * @remarks
-     * - Uses the configured video analysis model to generate a text analysis of the video.
-     * - The prompt should describe what to look for or extract from the video.
+     * - The prompt describes what to look for or extract from the video.
+     * - The default model is gemini-3.8-flash. Any Gemini chat model reads video (the 3.x Flash and Flash Lite models, gemini-3.1-pro, and the 2.5 family), as does TwelveLabs' pegasus-1.6; set one with `videoAnalysisModelOverride`, with `temperature` and `maxResponseTokens` beside `model`.
+     * - `videoUrl` is a video file's URL (an app file, a Google Drive share link, or any public URL). Gemini also reads a public YouTube link; Pegasus doesn't.
+     * - Length: Pegasus takes 1 second to 1 hour. Gemini takes about 3 hours at its default resolution; the 3.x models read a second of video as about 66 tokens and the 2.5 models as about 258, which is what a long video costs.
      *
      * @example
      * ```typescript
@@ -9744,9 +9750,10 @@ interface StepMethods {
      * Split an image into a base image and transparent layers.
      *
      * @remarks
-     * - Uses Seedream 5.0 Flash layer decomposition. There is no model picker.
-     * - An optional prompt can name the objects or groupings to separate.
-     * - The output is an array of hosted image URLs. PNG output keeps layer transparency.
+     * - Uses Seedream 5.0 Flash layer decomposition by default, and Seedream 5.0 Pro when type is "advanced", which separates more cleanly and costs more (and twice as much at 2K).
+     * - `resolution` is "1k", "1.5k" or "2k" (the default).
+     * - An optional prompt names the objects or groupings to separate. Left out, the model separates the main elements itself: text, subjects, decorations and background.
+     * - The output is an array of hosted PNG image URLs: first the base image at full size, then each separated element as a transparent layer cropped to the element. The layers come with no position, so placing one back over the base means finding where it sat. Flash returned 13 layers for one photo and Pro 16.
      *
      * @example
      * ```typescript
@@ -10139,8 +10146,12 @@ interface StepMethods {
      * Generate a 3D model using a 3D generation model.
      *
      * @remarks
-     * - Text-to-3D models use the prompt field.
-     * - Image-to-3D and multi-view models take image URLs through the selected model's configuration inputs.
+     * - The default model is wavespeed-tripo3d-h3.1 (Tripo v3.1), which makes a model from the prompt, from one `image`, or from `images`: 2–4 views of the same object, in order front, left, back, right.
+     * - Models and what they take, set in `threeDModelOverride.config`:
+     * - From text or an image: wavespeed-tripo3d-h3.1, wavespeed-meshy7.1 (also `images`, 1–4 views), wavespeed-hunyuan3d-v3 (an `image`, plus optional `back_image_url`, `left_image_url`, `right_image_url`), wavespeed-hunyuan-3d-v3.1-rapid, wavespeed-tripo3d-p2, wavespeed-rodin-v2.5 (`images`, 1–5).
+     * - From an image only: trellis-2, wavespeed-meshy7 (or `images`, 1–4 views), wavespeed-tripo3d-v2.5, seed3d-2.0, wavespeed-sam-3d-objects (with optional `mask_images`).
+     * - From views: wavespeed-hi3d-v3.0 (`front_image`, and optional `back_image`, `left_image`, `right_image`).
+     * - `$mode` in the config picks a model's mode (`text-to-3d`, `image-to-3d`, or its multi-view mode). Left out, `images` picks the multi-view mode and `image` picks image-to-3d; with neither, the model's default mode. A text mode needs the prompt; in the others the prompt is optional guidance. A prompt longer than the model reads is refused, not cut.
      * - The output is standardized as a GLB URL plus optional FBX/OBJ/USDZ, thumbnail, texture maps, and provider task ID.
      *
      * @example
@@ -10183,7 +10194,11 @@ interface StepMethods {
      * Generate an audio file from provided instructions (text) using a music model.
      *
      * @remarks
-     * - The text field contains the instructions (prompt) for the music generation.
+     * - `text` describes the song: genre, mood, instruments, tempo, vocals, and for Lyria its structure (`[Verse]`, `[Chorus]`, timestamps such as `[0:00-0:20] Intro`). A text longer than the model reads is refused, not cut.
+     * - The default model is elevenlabs-music_v2.5. Options go in `musicModelOverride.config`, and differ by model:
+     * - elevenlabs-music_v2.5, elevenlabs-music_v2: `musicLengthMs`, the song's length from 3,000 to 600,000 (the model chooses one from the text otherwise, and holds a length it's given), and `forceInstrumental: true` for no vocals.
+     * - lyria-3.5: songs of up to about three minutes, their length steered from the text; `image`, the URL of an image for the song to take its mood and subject from. lyria-3 makes 30-second clips, lyria-3-pro songs of up to about three minutes.
+     * - minimax-music-3.0: `lyrics` (written from the text when left out), `is_instrumental: true`, `bitrate`, `sample_rate`. minimax-music-2.5 requires `lyrics`.
      * - In foreground mode, the audio is displayed to the user. In background mode, the URL is saved to a variable.
      *
      * @example
@@ -10220,7 +10235,12 @@ interface StepMethods {
      * Generate a sound effect from a text description using a sound model.
      *
      * @remarks
-     * - The text field contains the description of the sound effect to generate.
+     * - `text` describes the sound effect to generate.
+     * - The default model is elevenlabs-sound_v2. Options go in `soundModelOverride.config`:
+     * - `durationSeconds` — the length, from 0.5 to 30. Left out, the model picks one from the text, and the effect is billed as 5 seconds whatever its length.
+     * - `loop: true` — a sound that repeats with no audible start or end, for ambience and backgrounds.
+     * - `promptInfluence` — how literally the text is followed, from 0 to 1 (default 0.3).
+     * - `outputFormat` — MP3 (`mp3_44100_128` by default, up to `mp3_44100_192`) or Opus (`opus_48000_32` to `opus_48000_192`).
      * - In foreground mode, the audio is displayed to the user. In background mode, the URL is saved to a variable.
      *
      * @example

package/dist/index.js

@@ -4169,7 +4169,7 @@ var stepMetadata = {
   "analyzeVideo": {
     stepType: "analyzeVideo",
     description: "Analyze a video using a video analysis model based on a text prompt.",
-    usageNotes: "- Uses the configured video analysis model to generate a text analysis of the video.\n- The prompt should describe what to look for or extract from the video.",
+    usageNotes: "- The prompt describes what to look for or extract from the video.\n- The default model is gemini-3.8-flash. Any Gemini chat model reads video (the 3.x Flash and Flash Lite models, gemini-3.1-pro, and the 2.5 family), as does TwelveLabs' pegasus-1.6; set one with `videoAnalysisModelOverride`, with `temperature` and `maxResponseTokens` beside `model`.\n- `videoUrl` is a video file's URL (an app file, a Google Drive share link, or any public URL). Gemini also reads a public YouTube link; Pegasus doesn't.\n- Length: Pegasus takes 1 second to 1 hour. Gemini takes about 3 hours at its default resolution; the 3.x models read a second of video as about 66 tokens and the 2.5 models as about 258, which is what a long video costs.",
     inputSchema: { "type": "object", "properties": { "prompt": { "type": "string", "description": "Instructions describing what to look for or extract from the video" }, "videoUrl": { "type": "string", "description": "URL of the video to analyze" }, "videoAnalysisModelOverride": { "anyOf": [{ "type": "object", "properties": { "model": { "type": "string" }, "config": { "type": "object", "properties": {}, "required": [] } }, "required": ["model"] }, { "type": "object", "properties": { "model": { "type": "string", "description": 'Model identifier (e.g. "gpt-4", "claude-3-opus")' }, "temperature": { "type": "number", "description": "Sampling temperature for the model (0-2)" }, "maxResponseTokens": { "type": "number", "description": "Maximum number of tokens in the model's response" }, "ignorePreamble": { "type": "boolean", "description": "Whether to skip the system preamble/instructions" }, "userMessagePreprocessor": { "type": "object", "properties": { "dataSource": { "type": "string", "description": "Data source identifier for the preprocessor" }, "messageTemplate": { "type": "string", "description": "Template string applied to user messages before sending to the model" }, "maxResults": { "type": "number", "description": "Maximum number of results to include from the data source" }, "enabled": { "type": "boolean", "description": "Whether the preprocessor is active" }, "shouldInherit": { "type": "boolean", "description": "Whether child steps should inherit this preprocessor configuration" } }, "description": "Preprocessor applied to user messages before sending to the model" }, "preamble": { "type": "string", "description": "System preamble/instructions for the model" }, "multiModelEnabled": { "type": "boolean", "description": "Whether multi-model candidate generation is enabled" }, "editResponseEnabled": { "type": "boolean", "description": "Whether the user can edit the model's response" }, "promptCaching": { "type": "boolean", "description": "Prompt caching, on models that bill cache writes. Absent: cached where something reads it back (the later steps of a run, a chat). `false` turns it off, for a single call such as a daily scheduled run. `true` caches even a single call, for a prompt resent unchanged within minutes." }, "config": { "type": "object", "properties": {}, "required": [], "description": "Additional model-specific configuration" } }, "required": ["model", "temperature", "maxResponseTokens"] }] } }, "required": ["prompt", "videoUrl"] },
     outputSchema: { "type": "object", "properties": { "analysis": { "type": "string", "description": "Text analysis of the video generated by the video analysis model" } }, "required": ["analysis"] }
   },
@@ -4269,8 +4269,8 @@ var stepMetadata = {
   "decomposeImage": {
     stepType: "decomposeImage",
     description: "Split an image into a base image and transparent layers.",
-    usageNotes: "- Uses Seedream 5.0 Flash layer decomposition. There is no model picker.\n- An optional prompt can name the objects or groupings to separate.\n- The output is an array of hosted image URLs. PNG output keeps layer transparency.",
-    inputSchema: { "type": "object", "properties": { "image": { "type": "string", "description": "Image to split into a base image and transparent layers" }, "prompt": { "type": "string", "description": "Optional guidance for which objects or groupings to separate" }, "intermediateAsset": { "type": "boolean", "description": "When true, assets are created but hidden from the user's gallery" } }, "required": ["image"] },
+    usageNotes: '- Uses Seedream 5.0 Flash layer decomposition by default, and Seedream 5.0 Pro when type is "advanced", which separates more cleanly and costs more (and twice as much at 2K).\n- `resolution` is "1k", "1.5k" or "2k" (the default).\n- An optional prompt names the objects or groupings to separate. Left out, the model separates the main elements itself: text, subjects, decorations and background.\n- The output is an array of hosted PNG image URLs: first the base image at full size, then each separated element as a transparent layer cropped to the element. The layers come with no position, so placing one back over the base means finding where it sat. Flash returned 13 layers for one photo and Pro 16.',
+    inputSchema: { "type": "object", "properties": { "image": { "type": "string", "description": "Image to split into a base image and transparent layers" }, "prompt": { "type": "string", "description": "Optional guidance for which objects or groupings to separate" }, "type": { "enum": ["standard", "advanced"], "type": "string", "description": 'Decomposition quality tier: "standard" (default) or "advanced"' }, "resolution": { "enum": ["1k", "1.5k", "2k"], "type": "string", "description": 'Output resolution: "1k", "1.5k" or "2k" (default)' }, "intermediateAsset": { "type": "boolean", "description": "When true, assets are created but hidden from the user's gallery" } }, "required": ["image"] },
     outputSchema: { "type": "object", "properties": { "imageUrls": { "type": "array", "items": { "type": "string" } } }, "required": ["imageUrls"] }
   },
   "deleteDataSource": {
@@ -4430,7 +4430,7 @@ var stepMetadata = {
   "generate3dModel": {
     stepType: "generate3dModel",
     description: "Generate a 3D model using a 3D generation model.",
-    usageNotes: "- Text-to-3D models use the prompt field.\n- Image-to-3D and multi-view models take image URLs through the selected model's configuration inputs.\n- The output is standardized as a GLB URL plus optional FBX/OBJ/USDZ, thumbnail, texture maps, and provider task ID.",
+    usageNotes: "- The default model is wavespeed-tripo3d-h3.1 (Tripo v3.1), which makes a model from the prompt, from one `image`, or from `images`: 2\u20134 views of the same object, in order front, left, back, right.\n- Models and what they take, set in `threeDModelOverride.config`:\n  - From text or an image: wavespeed-tripo3d-h3.1, wavespeed-meshy7.1 (also `images`, 1\u20134 views), wavespeed-hunyuan3d-v3 (an `image`, plus optional `back_image_url`, `left_image_url`, `right_image_url`), wavespeed-hunyuan-3d-v3.1-rapid, wavespeed-tripo3d-p2, wavespeed-rodin-v2.5 (`images`, 1\u20135).\n  - From an image only: trellis-2, wavespeed-meshy7 (or `images`, 1\u20134 views), wavespeed-tripo3d-v2.5, seed3d-2.0, wavespeed-sam-3d-objects (with optional `mask_images`).\n  - From views: wavespeed-hi3d-v3.0 (`front_image`, and optional `back_image`, `left_image`, `right_image`).\n- `$mode` in the config picks a model's mode (`text-to-3d`, `image-to-3d`, or its multi-view mode). Left out, `images` picks the multi-view mode and `image` picks image-to-3d; with neither, the model's default mode. A text mode needs the prompt; in the others the prompt is optional guidance. A prompt longer than the model reads is refused, not cut.\n- The output is standardized as a GLB URL plus optional FBX/OBJ/USDZ, thumbnail, texture maps, and provider task ID.",
     inputSchema: { "type": "object", "properties": { "prompt": { "type": "string", "description": "Text prompt for text-to-3D models, or optional guidance for image-to-3D models" }, "intermediateAsset": { "type": "boolean", "description": "When true, the asset is created but hidden from the user's gallery (tagged as intermediate)" }, "threeDModelOverride": { "type": "object", "properties": { "model": { "type": "string", "description": "3D generation model identifier" }, "config": { "type": "object", "properties": {}, "required": [], "description": "Additional model-specific configuration" } }, "required": ["model"], "description": "Optional model configuration override. Uses the workflow's default 3D model if not specified" } } },
     outputSchema: { "type": "object", "properties": { "prompt": { "type": "string" }, "glbUrl": { "type": "string" }, "fbxUrl": { "type": "string" }, "objUrl": { "type": "string" }, "usdzUrl": { "type": "string" }, "thumbnailUrl": { "type": "string" }, "textureUrls": { "type": "array", "items": { "type": "object", "properties": {}, "required": [] } }, "providerTaskId": { "type": "string" }, "resolvedConfig": { "type": "object", "properties": {}, "required": [] } }, "required": ["glbUrl"] }
   },
@@ -4458,7 +4458,7 @@ var stepMetadata = {
   "generateMusic": {
     stepType: "generateMusic",
     description: "Generate an audio file from provided instructions (text) using a music model.",
-    usageNotes: "- The text field contains the instructions (prompt) for the music generation.\n- In foreground mode, the audio is displayed to the user. In background mode, the URL is saved to a variable.",
+    usageNotes: "- `text` describes the song: genre, mood, instruments, tempo, vocals, and for Lyria its structure (`[Verse]`, `[Chorus]`, timestamps such as `[0:00-0:20] Intro`). A text longer than the model reads is refused, not cut.\n- The default model is elevenlabs-music_v2.5. Options go in `musicModelOverride.config`, and differ by model:\n  - elevenlabs-music_v2.5, elevenlabs-music_v2: `musicLengthMs`, the song's length from 3,000 to 600,000 (the model chooses one from the text otherwise, and holds a length it's given), and `forceInstrumental: true` for no vocals.\n  - lyria-3.5: songs of up to about three minutes, their length steered from the text; `image`, the URL of an image for the song to take its mood and subject from. lyria-3 makes 30-second clips, lyria-3-pro songs of up to about three minutes.\n  - minimax-music-3.0: `lyrics` (written from the text when left out), `is_instrumental: true`, `bitrate`, `sample_rate`. minimax-music-2.5 requires `lyrics`.\n- In foreground mode, the audio is displayed to the user. In background mode, the URL is saved to a variable.",
     inputSchema: { "type": "object", "properties": { "text": { "type": "string", "description": "The instructions (prompt) for the music generation" }, "intermediateAsset": { "type": "boolean", "description": "When true, the asset is created but hidden from the user's gallery (tagged as intermediate)" }, "musicModelOverride": { "type": "object", "properties": { "model": { "type": "string" }, "config": { "type": "object", "properties": {}, "required": [] } }, "required": ["model"], "description": "Optional model configuration override. Uses the workflow's default music model if not specified" } }, "required": ["text"] },
     outputSchema: { "type": "object", "properties": { "audioUrl": { "type": "string", "description": "URL of the generated audio file" } }, "required": ["audioUrl"] }
   },
@@ -4472,7 +4472,7 @@ var stepMetadata = {
   "generateSound": {
     stepType: "generateSound",
     description: "Generate a sound effect from a text description using a sound model.",
-    usageNotes: "- The text field contains the description of the sound effect to generate.\n- In foreground mode, the audio is displayed to the user. In background mode, the URL is saved to a variable.",
+    usageNotes: "- `text` describes the sound effect to generate.\n- The default model is elevenlabs-sound_v2. Options go in `soundModelOverride.config`:\n  - `durationSeconds` \u2014 the length, from 0.5 to 30. Left out, the model picks one from the text, and the effect is billed as 5 seconds whatever its length.\n  - `loop: true` \u2014 a sound that repeats with no audible start or end, for ambience and backgrounds.\n  - `promptInfluence` \u2014 how literally the text is followed, from 0 to 1 (default 0.3).\n  - `outputFormat` \u2014 MP3 (`mp3_44100_128` by default, up to `mp3_44100_192`) or Opus (`opus_48000_32` to `opus_48000_192`).\n- In foreground mode, the audio is displayed to the user. In background mode, the URL is saved to a variable.",
     inputSchema: { "type": "object", "properties": { "text": { "type": "string", "description": "The description of the sound effect to generate" }, "intermediateAsset": { "type": "boolean", "description": "When true, the asset is created but hidden from the user's gallery (tagged as intermediate)" }, "soundModelOverride": { "type": "object", "properties": { "model": { "type": "string" }, "config": { "type": "object", "properties": {}, "required": [] } }, "required": ["model"], "description": "Optional model configuration override. Uses the workflow's default sound model if not specified" } }, "required": ["text"] },
     outputSchema: { "type": "object", "properties": { "audioUrl": { "type": "string", "description": "URL of the generated audio file" } }, "required": ["audioUrl"] }
   },
@@ -4761,7 +4761,7 @@ var stepMetadata = {
     stepType: "meshyTextTo3d",
     description: "Generate a 3D model preview from a text prompt using Meshy. Produces an untextured mesh (preview stage).",
     usageNotes: "- Creates a text-to-3D preview task (mesh generation only, no texture).\n- Use the Meshy Texture step to apply textures to the preview.\n- Maximum prompt length is 600 characters.",
-    inputSchema: { "type": "object", "properties": { "prompt": { "type": "string", "description": "Description of the 3D model to generate (max 600 characters)" }, "modelType": { "type": "string", "description": '"standard" (default) or "lowpoly". Lowpoly ignores topology/target_polycount.' }, "topology": { "type": "string", "description": '"triangle" (default) or "quad"' }, "targetPolycount": { "type": "number", "description": "Target polygon count (default 30000, range 100-300000)" }, "symmetryMode": { "type": "string", "description": 'Symmetry mode: "auto" (default), "off", or "on"' }, "poseMode": { "type": "string", "description": 'Pose mode: "a-pose", "t-pose", or "" (default, no specific pose)' } }, "required": ["prompt"] },
+    inputSchema: { "type": "object", "properties": { "prompt": { "type": "string", "description": "Description of the 3D model to generate (max 600 characters)" }, "modelType": { "type": "string", "description": '"standard" (default) or "smart-topology", clean low-poly geometry (target_polycount 100-15000). "lowpoly", which Meshy retires on 2026-10-30, runs as smart-topology.' }, "topology": { "type": "string", "description": '"triangle" (default) or "quad"' }, "targetPolycount": { "type": "number", "description": "Target polygon count (default 30000, range 100-300000)" }, "symmetryMode": { "type": "string", "description": 'Symmetry mode: "auto" (default), "off", or "on"' }, "poseMode": { "type": "string", "description": 'Pose mode: "a-pose", "t-pose", or "" (default, no specific pose)' } }, "required": ["prompt"] },
     outputSchema: { "type": "object", "properties": { "glbUrl": { "type": "string" }, "fbxUrl": { "type": "string" }, "objUrl": { "type": "string" }, "usdzUrl": { "type": "string" }, "thumbnailUrl": { "type": "string" }, "textureUrls": { "type": "array", "items": { "type": "object", "properties": {}, "required": [] } }, "animations": { "type": "array", "items": { "type": "object", "properties": { "name": { "type": "string" }, "glbUrl": { "type": "string" }, "fbxUrl": { "type": "string" } }, "required": ["name"] } }, "providerTaskId": { "type": "string" } }, "required": ["glbUrl"] }
   },
   "meshyTexture": {