@mindstudio-ai/agent 0.1.181 → 0.1.182

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/llms.txt CHANGED
@@ -356,6 +356,14 @@ Legacy v1 data sources. v2 apps should use `dataSources.defineDataSource()` from
356
356
  - Input: `{ dataSourceId: string, documentId: string }`
357
357
  - Output: `{ dataSourceId: string, documentId: string }`
358
358
 
359
+ #### describeExternalModel
360
+ Read what a community model on Replicate takes, returns and costs, before running it
361
+ - Returns the model pinned to a version (`model`, to pass to `runExternalModel` as it is), its input and output schemas, its licence and its price per second of run time
362
+ - Free. Use it while building, to learn a model's field names, and keep the pinned `model` in the app's code: a model's latest version can change its fields
363
+ - Models Replicate maintains itself ("official" models) and private models are refused
364
+ - Input: `{ replicate: { model: string } }`
365
+ - Output: `{ model: string, url: string, description: string, inputSchema: object, outputSchema: object, licenseUrl?: string, pricePerSecond: number, priceIsUpperBound: boolean }`
366
+
359
367
  #### detectChanges
360
368
  Detect changes between runs by comparing current input against previously stored state. Routes execution based on whether a change occurred.
361
369
  - Persists state across runs using a global variable keyed to the step ID.
@@ -420,8 +428,8 @@ Ask structured questions about text, JSON, or images and get calibrated answers
420
428
  - images is for a model that reads them. Pass URLs or data URLs. A model that reads none refuses the call.
421
429
  - The response model field is the version that answered.
422
430
  - Billed per input token. Output tokens are free. Each model has its own rate, and GPT-6 Luna Decisions bills a prompt over 272k tokens at twice its rate.
423
- - Input: `{ state: string | object | unknown[], questions: object, images?: string[], decisionModelOverride?: { model: string, config?: object } }`
424
- - Output: `{ model: string, answers: object, usage: { input_tokens: number, output_tokens: number } }`
431
+ - Input: `{ state: string | object | unknown[], questions: Record<string, { type: string, instructions: string | object | unknown[], criteria?: { true?: string, false?: string } } | { type: string, instructions: string | object | unknown[], criteria: Record<string, string | null> } | { type: string, instructions: string | object | unknown[], criteria: string[] }>, images?: string[], decisionModelOverride?: { model: string, config?: object } }`
432
+ - Output: `{ model: string, answers: Record<string, { type: string, noul: number } | { type: string, choice: string, confidence: number, probabilities: Record<string, number> } | { type: string, score: number, confidence: number, legend: Record<string, string>, probabilities: Record<string, number> } | { type: string }>, usage: { input_tokens: number, output_tokens: number } }`
425
433
 
426
434
  #### extractAudioFromVideo
427
435
  Extract audio MP3 from a video file
@@ -470,7 +478,7 @@ Generate a 3D model using a 3D generation model.
470
478
  - `$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.
471
479
  - The output is standardized as a GLB URL plus optional FBX/OBJ/USDZ, thumbnail, texture maps, and provider task ID.
472
480
  - Input: `{ prompt?: string, intermediateAsset?: boolean, threeDModelOverride?: { model: string, config?: object } }`
473
- - Output: `{ prompt?: string, glbUrl: string, fbxUrl?: string, objUrl?: string, usdzUrl?: string, thumbnailUrl?: string, textureUrls?: object[], providerTaskId?: string, resolvedConfig?: object }`
481
+ - Output: `{ prompt?: string, glbUrl: string, fbxUrl?: string, objUrl?: string, usdzUrl?: string, thumbnailUrl?: string, textureUrls?: Record<string, string>[], providerTaskId?: string, resolvedConfig?: object }`
474
482
 
475
483
  #### generateAsset
476
484
  Generate an HTML asset and export it as a webpage, PDF, or image
@@ -479,7 +487,7 @@ Generate an HTML asset and export it as a webpage, PDF, or image
479
487
  - If PDF or composited image generation are part of the workflow, assistant adds the block and leaves the "source" empty. In a separate step, assistant generates a detailed request for the developer who will write the HTML.
480
488
  - Can also auto-generate HTML from a prompt (like a generate text block to generate HTML). In these cases, create a prompt with variables in the dynamicPrompt variable describing, in detail, the document to generate
481
489
  - Can either display output directly to user (foreground mode) or save the URL of the asset to a variable (background mode)
482
- - Input: `{ source: string, sourceType: "html" | "markdown" | "spa" | "raw" | "dynamic" | "customInterface", outputFormat: "pdf" | "png" | "html" | "mp4" | "openGraph", pageSize: "full" | "letter" | "A4" | "custom", testData: object, options?: { pageWidthPx?: number, pageHeightPx?: number, pageOrientation?: "portrait" | "landscape", rehostMedia?: boolean, videoDurationSeconds?: number }, spaSource?: { source?: string, lastCompiledSource?: string, files?: object, paths: string[], root: string, zipUrl: string }, rawSource?: string, dynamicPrompt?: string, dynamicSourceModelOverride?: { model: string, temperature: number, maxResponseTokens: number, ignorePreamble?: boolean, userMessagePreprocessor?: { dataSource?: string, messageTemplate?: string, maxResults?: number, enabled?: boolean, shouldInherit?: boolean }, preamble?: string, multiModelEnabled?: boolean, editResponseEnabled?: boolean, promptCaching?: boolean, config?: object }, transitionControl?: "default" | "native", shareControl?: "default" | "hidden", shareImageUrl?: string, intermediateAsset?: boolean }`
490
+ - Input: `{ source: string, sourceType: "html" | "markdown" | "spa" | "raw" | "dynamic" | "customInterface", outputFormat: "pdf" | "png" | "html" | "mp4" | "openGraph", pageSize: "full" | "letter" | "A4" | "custom", testData: object, options?: { pageWidthPx?: number, pageHeightPx?: number, pageOrientation?: "portrait" | "landscape", rehostMedia?: boolean, videoDurationSeconds?: number }, spaSource?: { source?: string, lastCompiledSource?: string, files?: Record<string, { value: string }>, paths: string[], root: string, zipUrl: string }, rawSource?: string, dynamicPrompt?: string, dynamicSourceModelOverride?: { model: string, temperature: number, maxResponseTokens: number, ignorePreamble?: boolean, userMessagePreprocessor?: { dataSource?: string, messageTemplate?: string, maxResults?: number, enabled?: boolean, shouldInherit?: boolean }, preamble?: string, multiModelEnabled?: boolean, editResponseEnabled?: boolean, promptCaching?: boolean, config?: object }, transitionControl?: "default" | "native", shareControl?: "default" | "hidden", shareImageUrl?: string, intermediateAsset?: boolean }`
483
491
  - Output: `{ url: string }`
484
492
 
485
493
  #### generateImage
@@ -644,12 +652,12 @@ Trigger a Make.com (formerly Integromat) scenario via webhook and return the res
644
652
  - The webhook URL must be configured in your Make.com scenario.
645
653
  - Input key-value pairs are sent as JSON in the POST body.
646
654
  - Response format depends on the Make.com scenario configuration.
647
- - Input: `{ webhookUrl: string, input: object }`
655
+ - Input: `{ webhookUrl: string, input: Record<string, string> }`
648
656
  - Output: `{ data: unknown }`
649
657
 
650
658
  #### mergeAudio
651
659
  Merge one or more clips into a single audio file.
652
- - Input: `{ mp3Urls: string[], fileMetadata?: object, albumArtUrl?: string, intermediateAsset?: boolean }`
660
+ - Input: `{ mp3Urls: string[], fileMetadata?: Record<string, string>, albumArtUrl?: string, intermediateAsset?: boolean }`
653
661
  - Output: `{ audioUrl: string }`
654
662
 
655
663
  #### mergeVideos
@@ -672,7 +680,7 @@ Trigger an n8n workflow node via webhook and return the response.
672
680
  - The webhook URL must be configured in your n8n workflow.
673
681
  - Supports GET and POST methods with optional Basic authentication.
674
682
  - For GET requests, input values are sent as query parameters. For POST, they are sent as JSON body.
675
- - Input: `{ method: string, authentication: "none" | "basic" | "string", user: string, password: string, webhookUrl: string, input: object }`
683
+ - Input: `{ method: string, authentication: "none" | "basic" | "string", user: string, password: string, webhookUrl: string, input: Record<string, string> }`
676
684
  - Output: `{ data: unknown }`
677
685
 
678
686
  #### postToSlackChannel
@@ -688,7 +696,7 @@ Send data to a Zapier Zap via webhook and return the response.
688
696
  - The webhook URL must be configured in the Zapier Zap settings
689
697
  - Input keys and values are sent as the JSON body of the POST request
690
698
  - The webhook response (JSON or plain text) is returned as the output
691
- - Input: `{ webhookUrl: string, input: object }`
699
+ - Input: `{ webhookUrl: string, input: Record<string, string> }`
692
700
  - Output: `{ data: unknown }`
693
701
 
694
702
  #### queryDataSource
@@ -730,7 +738,7 @@ Render an HTML composition (HTML, CSS, media and seekable animation on a timelin
730
738
  - fps 24, 30 or 60 (default 30); quality draft, standard or high (default standard); format mp4 (default), webm or mov (both keep transparency), or gif
731
739
  - An mp4, webm or mov renders in chunks of about 10 seconds side by side, so a long video takes little longer than a short one; a call finishes within about 30 minutes, and reports its progress as it goes. A gif renders whole, within 15 minutes
732
740
  - A composition HyperFrames can't render fails with HyperFrames' own reason
733
- - Input: `{ html?: string, project?: string, entryFile?: string, variables?: object, fps?: 24 | 30 | 60, quality?: "draft" | "standard" | "high", format?: "mp4" | "webm" | "mov" | "gif", intermediateAsset?: boolean }`
741
+ - Input: `{ html?: string, project?: string, entryFile?: string, variables?: Record<string, unknown>, fps?: 24 | 30 | 60, quality?: "draft" | "standard" | "high", format?: "mp4" | "webm" | "mov" | "gif", intermediateAsset?: boolean }`
734
742
  - Output: `{ videoUrl: string, durationSec: number, frames: number }`
735
743
 
736
744
  #### resizeVideo
@@ -739,6 +747,18 @@ Resize a video file
739
747
  - Input: `{ videoUrl: string, mode: "fit" | "exact", maxWidth?: number, maxHeight?: number, width?: number, height?: number, strategy?: "pad" | "crop", intermediateAsset?: boolean }`
740
748
  - Output: `{ videoUrl: string }`
741
749
 
750
+ #### runExternalModel
751
+ Run a public community model the model catalog doesn't carry, on Replicate
752
+ - For a kind of model no other action covers (separating a song's stems, aligning words to audio, depth or segmentation maps, a published fine-tune). A model the catalog has is run with its own action
753
+ - `replicate.model` is `owner/name:version`. Call `describeExternalModel` first: it returns the model pinned to its latest version, its input and output schemas, its licence and its price
754
+ - `replicate.input` is the model's own input, passed through as given. A file input is a URL; a private file of the app's own is signed for the run
755
+ - `output` is whatever the model returns. Its files are copied before Replicate deletes them, an hour after the run: into the store passed as `store`, or onto the public CDN without one, so pass a private store for anything confidential
756
+ - Billed for the seconds the model ran (`runSeconds`), at the model's own per-second price. A cold start is waited for and not billed, and can take minutes on a model that hasn't run lately: `timeoutSeconds` covers the wait and the run together
757
+ - A run that fails throws with the model's own error. One that passes `timeoutSeconds` is cancelled and throws `external_model_timed_out`
758
+ - Models Replicate maintains itself ("official" models) and private models are refused. Each model's licence is its author's: `describeExternalModel` links it
759
+ - Input: `{ replicate: { model: string, input: object, timeoutSeconds?: number }, intermediateAsset?: boolean }`
760
+ - Output: `{ output: unknown, runSeconds: number }`
761
+
742
762
  #### runFfmpeg
743
763
  Run an ffmpeg command on media files
744
764
  - Runs one ffmpeg command, its arguments exactly as given, off the app's own sandbox (ffmpeg 9, a static build with the common codecs and filters)
@@ -747,8 +767,8 @@ Run an ffmpeg command on media files
747
767
  - One command per call, finishing within 15 minutes; up to 10 GB of inputs and outputs together, 5 GB per output
748
768
  - The system fonts are Noto Sans (`drawtext`'s default), DejaVu Sans, Liberation Sans and Noto Color Emoji: pass a font file as an input for any other
749
769
  - A command ffmpeg refuses fails with ffmpeg's own error lines
750
- - Input: `{ inputs: object, args: string[], intermediateAsset?: boolean }`
751
- - Output: `{ outputs: object }`
770
+ - Input: `{ inputs: Record<string, string>, args: string[], intermediateAsset?: boolean }`
771
+ - Output: `{ outputs: Record<string, string> }`
752
772
 
753
773
  #### runFromConnectorRegistry
754
774
  Run a raw API connector to a third-party service
@@ -757,7 +777,7 @@ Run a raw API connector to a third-party service
757
777
  - Use /developer/v2/helpers/connections to list your available OAuth connections.
758
778
  - The actionId format is "serviceId/actionId" (e.g., "slack/send-message").
759
779
  - Pass a __connectionId to authenticate the request with a specific OAuth connection, otherwise the default will be used (if configured).
760
- - Input: `{ actionId: string, displayName: string, icon: string, configurationValues: object, __connectionId?: string }`
780
+ - Input: `{ actionId: string, displayName: string, icon: string, configurationValues: Record<string, string>, __connectionId?: string }`
761
781
  - Output: `{ data: object }`
762
782
 
763
783
  #### scrapeLinkedInCompany
@@ -780,7 +800,7 @@ Extract text, HTML, or structured content from one or more web pages.
780
800
  - Output format controls the result shape: "text" returns markdown, "html" returns raw HTML, "json" returns structured scraper data, "summary" returns a model-written summary and requires the "firecrawl" service.
781
801
  - Can optionally capture a screenshot of each page. Screenshots go to the public CDN by default, or into one of the app's own file stores when the call passes `store`.
782
802
  - Handles bot protection automatically; no proxy or rendering configuration is needed.
783
- - Input: `{ url: string, service?: "default" | "firecrawl", autoEnhance?: boolean, outputFormat?: "text" | "json" | "html" | "summary", pageOptions?: { onlyMainContent: boolean, screenshot: boolean, waitFor: number, replaceAllPathsWithAbsolutePaths: boolean, headers: object, removeTags: string[], mobile: boolean } }`
803
+ - Input: `{ url: string, service?: "default" | "firecrawl", autoEnhance?: boolean, outputFormat?: "text" | "json" | "html" | "summary", pageOptions?: { onlyMainContent: boolean, screenshot: boolean, waitFor: number, replaceAllPathsWithAbsolutePaths: boolean, headers: Record<string, string>, removeTags: string[], mobile: boolean } }`
784
804
  - Output: `{ content: string | string[] | { text: string, html: string, json?: object, screenshotUrl?: string, metadata?: { title: string, description: string, url: string, image: string }, error?: { code: string, message: string }, costUnits?: number } | { text: string, html: string, json?: object, screenshotUrl?: string, metadata?: { title: string, description: string, url: string, image: string }, error?: { code: string, message: string }, costUnits?: number }[], screenshot?: string | string[] }`
785
805
 
786
806
  #### scrapeXPost
@@ -848,7 +868,7 @@ Fetch Google Trends data for a search term.
848
868
  - date accepts shorthand ("now 1-H", "today 1-m", "today 5-y", etc.) or custom "yyyy-mm-dd yyyy-mm-dd" ranges.
849
869
  - data_type controls the shape of returned data: TIMESERIES, GEO_MAP, GEO_MAP_0, RELATED_TOPICS, or RELATED_QUERIES.
850
870
  - Input: `{ text: string, hl: string, geo: string, data_type: "TIMESERIES" | "GEO_MAP" | "GEO_MAP_0" | "RELATED_TOPICS" | "RELATED_QUERIES", cat: string, date: string, ts: string }`
851
- - Output: `{ trends: object }`
871
+ - Output: `{ trends: Record<string, unknown> }`
852
872
 
853
873
  #### searchPerplexity
854
874
  Search the web using the Perplexity API and return structured results.
@@ -1045,7 +1065,7 @@ Create or sync a contact in ActiveCampaign.
1045
1065
  - Requires an ActiveCampaign OAuth connection (connectionId).
1046
1066
  - If a contact with the email already exists, it may be updated depending on ActiveCampaign settings.
1047
1067
  - Custom fields are passed as a key-value map where keys are field IDs.
1048
- - Input: `{ email: string, firstName: string, lastName: string, phone: string, accountId: string, customFields: object, connectionId?: string }`
1068
+ - Input: `{ email: string, firstName: string, lastName: string, phone: string, accountId: string, customFields: Record<string, string>, connectionId?: string }`
1049
1069
  - Output: `{ contactId: string }`
1050
1070
 
1051
1071
  ### Airtable
@@ -1070,7 +1090,7 @@ Fetch a single record from an Airtable table by its record ID.
1070
1090
  - Requires an active Airtable OAuth connection (connectionId).
1071
1091
  - If the record is not found, returns a string message instead of a record object.
1072
1092
  - Input: `{ connectionId?: string, baseId: string, tableId: string, recordId: string }`
1073
- - Output: `{ record: { id: string, createdTime: string, fields: object } | null }`
1093
+ - Output: `{ record: { id: string, createdTime: string, fields: Record<string, unknown> } | null }`
1074
1094
 
1075
1095
  #### airtableGetTableRecords
1076
1096
  Fetch multiple records from an Airtable table with optional pagination.
@@ -1078,7 +1098,7 @@ Fetch multiple records from an Airtable table with optional pagination.
1078
1098
  - Default limit is 100 records. Maximum is 1000.
1079
1099
  - When outputFormat is 'csv', the variable receives CSV text. The direct execution output always returns parsed records.
1080
1100
  - Input: `{ connectionId?: string, baseId: string, tableId: string, outputFormat?: "json" | "csv", limit?: number }`
1081
- - Output: `{ records: { id: string, createdTime: string, fields: object }[] }`
1101
+ - Output: `{ records: { id: string, createdTime: string, fields: Record<string, unknown> }[] }`
1082
1102
 
1083
1103
  ### Apollo
1084
1104
 
@@ -1123,7 +1143,7 @@ Search for a row in a Coda table by matching column values.
1123
1143
  - Returns the first row matching all specified column values, or null if no match.
1124
1144
  - Search criteria in rowData are ANDed together.
1125
1145
  - Input: `{ connectionId?: string, docId: string, tableId: string, rowData: object }`
1126
- - Output: `{ row: { id: string, values: object } | null }`
1146
+ - Output: `{ row: { id: string, values: Record<string, unknown> } | null }`
1127
1147
 
1128
1148
  #### codaGetPage
1129
1149
  Export and read the contents of a page from a Coda document.
@@ -1139,7 +1159,7 @@ Fetch rows from a Coda table with optional pagination.
1139
1159
  - Default limit is 10000 rows. Rows are fetched in pages of 500.
1140
1160
  - When outputFormat is 'csv', the variable receives CSV text. The direct execution output always returns parsed rows.
1141
1161
  - Input: `{ connectionId?: string, docId: string, tableId: string, limit?: number | string, outputFormat?: "json" | "csv" }`
1142
- - Output: `{ rows: { id: string, values: object }[] }`
1162
+ - Output: `{ rows: { id: string, values: Record<string, unknown> }[] }`
1143
1163
 
1144
1164
  ### Gmail
1145
1165
 
@@ -1336,7 +1356,7 @@ Look up a HubSpot company by domain name or company ID.
1336
1356
  - When searching by domain, performs a search query then fetches the full company record.
1337
1357
  - Use additionalProperties to request specific HubSpot properties beyond the defaults.
1338
1358
  - Input: `{ connectionId?: string, searchBy: "domain" | "id", companyDomain: string, companyId: string, additionalProperties: string[] }`
1339
- - Output: `{ company: { id: string, properties: object, createdAt: string, updatedAt: string, archived: boolean } | null }`
1359
+ - Output: `{ company: { id: string, properties: Record<string, string | null>, createdAt: string, updatedAt: string, archived: boolean } | null }`
1340
1360
 
1341
1361
  #### hubspotGetContact
1342
1362
  Look up a HubSpot contact by email address or contact ID.
@@ -1344,7 +1364,7 @@ Look up a HubSpot contact by email address or contact ID.
1344
1364
  - Returns null if the contact is not found.
1345
1365
  - Use additionalProperties to request specific HubSpot properties beyond the defaults.
1346
1366
  - Input: `{ connectionId?: string, searchBy: "email" | "id", contactEmail: string, contactId: string, additionalProperties: string[] }`
1347
- - Output: `{ contact: { id: string, properties: object, createdAt: string, updatedAt: string, archived: boolean } | null }`
1367
+ - Output: `{ contact: { id: string, properties: Record<string, string | null>, createdAt: string, updatedAt: string, archived: boolean } | null }`
1348
1368
 
1349
1369
  ### Hunter.io
1350
1370
 
@@ -1408,7 +1428,7 @@ Apply a preset animation to a rigged 3D character model using Meshy.
1408
1428
  - Supports post-processing: FPS change (24/25/30/60), FBX-to-USDZ conversion, or armature extraction.
1409
1429
  - Animation categories: DailyActions, WalkAndRun, Fighting, Dancing, BodyMovements.
1410
1430
  - Input: `{ rigTaskId: string, actionId: number }`
1411
- - Output: `{ glbUrl: string, fbxUrl?: string, objUrl?: string, usdzUrl?: string, thumbnailUrl?: string, textureUrls?: object[], animations?: { name: string, glbUrl?: string, fbxUrl?: string }[], providerTaskId?: string }`
1431
+ - Output: `{ glbUrl: string, fbxUrl?: string, objUrl?: string, usdzUrl?: string, thumbnailUrl?: string, textureUrls?: Record<string, string>[], animations?: { name: string, glbUrl?: string, fbxUrl?: string }[], providerTaskId?: string }`
1412
1432
 
1413
1433
  #### meshyImageTo3d
1414
1434
  Generate a 3D model from one or more images using Meshy. Uses the multi-image-to-3D endpoint.
@@ -1416,7 +1436,7 @@ Generate a 3D model from one or more images using Meshy. Uses the multi-image-to
1416
1436
  - By default generates with textures. Set shouldTexture to false for mesh-only output.
1417
1437
  - Uses should_remesh: false to preserve UV mapping integrity.
1418
1438
  - Input: `{ imageUrls: string[], shouldTexture?: boolean, topology?: string, targetPolycount?: number, symmetryMode?: string, poseMode?: string, textureImageUrl?: string }`
1419
- - Output: `{ glbUrl: string, fbxUrl?: string, objUrl?: string, usdzUrl?: string, thumbnailUrl?: string, textureUrls?: object[], animations?: { name: string, glbUrl?: string, fbxUrl?: string }[], providerTaskId?: string }`
1439
+ - Output: `{ glbUrl: string, fbxUrl?: string, objUrl?: string, usdzUrl?: string, thumbnailUrl?: string, textureUrls?: Record<string, string>[], animations?: { name: string, glbUrl?: string, fbxUrl?: string }[], providerTaskId?: string }`
1420
1440
 
1421
1441
  #### meshyRemesh
1422
1442
  Remesh an existing 3D model to adjust topology, polygon count, or convert formats using Meshy.
@@ -1424,7 +1444,7 @@ Remesh an existing 3D model to adjust topology, polygon count, or convert format
1424
1444
  - Defaults to triangle topology with 30,000 target polys.
1425
1445
  - Useful for reducing face count before rigging (max 300k faces for rigging).
1426
1446
  - Input: `{ inputTaskId?: string, modelUrl?: string, topology?: string, targetPolycount?: number, resizeHeight?: number }`
1427
- - Output: `{ glbUrl: string, fbxUrl?: string, objUrl?: string, usdzUrl?: string, thumbnailUrl?: string, textureUrls?: object[], animations?: { name: string, glbUrl?: string, fbxUrl?: string }[], providerTaskId?: string }`
1447
+ - Output: `{ glbUrl: string, fbxUrl?: string, objUrl?: string, usdzUrl?: string, thumbnailUrl?: string, textureUrls?: Record<string, string>[], animations?: { name: string, glbUrl?: string, fbxUrl?: string }[], providerTaskId?: string }`
1428
1448
 
1429
1449
  #### meshyRig
1430
1450
  Auto-rig a humanoid 3D model and generate basic walking/running animations using Meshy.
@@ -1433,7 +1453,7 @@ Auto-rig a humanoid 3D model and generate basic walking/running animations using
1433
1453
  - Models with more than 300,000 faces should be remeshed first.
1434
1454
  - Returns rigged model files and optional basic animations.
1435
1455
  - Input: `{ inputTaskId?: string, modelUrl?: string, heightMeters?: number }`
1436
- - Output: `{ glbUrl: string, fbxUrl?: string, objUrl?: string, usdzUrl?: string, thumbnailUrl?: string, textureUrls?: object[], animations?: { name: string, glbUrl?: string, fbxUrl?: string }[], providerTaskId?: string }`
1456
+ - Output: `{ glbUrl: string, fbxUrl?: string, objUrl?: string, usdzUrl?: string, thumbnailUrl?: string, textureUrls?: Record<string, string>[], animations?: { name: string, glbUrl?: string, fbxUrl?: string }[], providerTaskId?: string }`
1437
1457
 
1438
1458
  #### meshyTextTo3d
1439
1459
  Generate a 3D model preview from a text prompt using Meshy. Produces an untextured mesh (preview stage).
@@ -1441,7 +1461,7 @@ Generate a 3D model preview from a text prompt using Meshy. Produces an untextur
1441
1461
  - Use the Meshy Texture step to apply textures to the preview.
1442
1462
  - Maximum prompt length is 600 characters.
1443
1463
  - Input: `{ prompt: string, modelType?: string, topology?: string, targetPolycount?: number, symmetryMode?: string, poseMode?: string }`
1444
- - Output: `{ glbUrl: string, fbxUrl?: string, objUrl?: string, usdzUrl?: string, thumbnailUrl?: string, textureUrls?: object[], animations?: { name: string, glbUrl?: string, fbxUrl?: string }[], providerTaskId?: string }`
1464
+ - Output: `{ glbUrl: string, fbxUrl?: string, objUrl?: string, usdzUrl?: string, thumbnailUrl?: string, textureUrls?: Record<string, string>[], animations?: { name: string, glbUrl?: string, fbxUrl?: string }[], providerTaskId?: string }`
1445
1465
 
1446
1466
  #### meshyTexture
1447
1467
  Apply or replace textures on a 3D model using a text prompt or reference image via Meshy.
@@ -1451,7 +1471,7 @@ Apply or replace textures on a 3D model using a text prompt or reference image v
1451
1471
  - By default preserves original UVs (enableOriginalUv = true).
1452
1472
  - Works with any model source: text-to-3D previews, image-to-3D, remeshed models, or external files.
1453
1473
  - Input: `{ inputTaskId?: string, modelUrl?: string, textStylePrompt?: string, imageStyleUrl?: string, enableOriginalUv?: boolean, enablePbr?: boolean }`
1454
- - Output: `{ glbUrl: string, fbxUrl?: string, objUrl?: string, usdzUrl?: string, thumbnailUrl?: string, textureUrls?: object[], animations?: { name: string, glbUrl?: string, fbxUrl?: string }[], providerTaskId?: string }`
1474
+ - Output: `{ glbUrl: string, fbxUrl?: string, objUrl?: string, usdzUrl?: string, thumbnailUrl?: string, textureUrls?: Record<string, string>[], animations?: { name: string, glbUrl?: string, fbxUrl?: string }[], providerTaskId?: string }`
1455
1475
 
1456
1476
  ### Notion
1457
1477
 
@@ -1479,7 +1499,7 @@ Call any QuickBooks Online Accounting API route using a connected company.
1479
1499
  - Route is relative to /v3/company/{realmId}/, e.g. "account", "account/93", "query", "invoice".
1480
1500
  - Use queryParams for query-string values such as query=select * from Account.
1481
1501
  - Body is JSON for POST, PUT, and PATCH.
1482
- - Input: `{ connectionId?: string, method: string, route: string, queryParams: object, body: string }`
1502
+ - Input: `{ connectionId?: string, method: string, route: string, queryParams: Record<string, string>, body: string }`
1483
1503
  - Output: `{ ok: boolean, status: number, body: unknown }`
1484
1504
 
1485
1505
  ### TypeSafe
@@ -1487,8 +1507,8 @@ Call any QuickBooks Online Accounting API route using a connected company.
1487
1507
  #### typesafeEvaluate
1488
1508
  Use evaluateDecision: the same questions, answers and models, with decisionModelOverride in place of model. evaluateDecision's earlier name, which still runs: ask structured questions about text, JSON, or images and get calibrated answers from a decision model.
1489
1509
  - Runs exactly as evaluateDecision does, with the same questions, answers, models and limits (see its notes), and model in place of decisionModelOverride.model.
1490
- - Input: `{ state: string | object | unknown[], questions: object, model?: string, images?: string[] }`
1491
- - Output: `{ model: string, answers: object, usage: { input_tokens: number, output_tokens: number } }`
1510
+ - Input: `{ state: string | object | unknown[], questions: Record<string, { type: string, instructions: string | object | unknown[], criteria?: { true?: string, false?: string } } | { type: string, instructions: string | object | unknown[], criteria: Record<string, string | null> } | { type: string, instructions: string | object | unknown[], criteria: string[] }>, model?: string, images?: string[] }`
1511
+ - Output: `{ model: string, answers: Record<string, { type: string, noul: number } | { type: string, choice: string, confidence: number, probabilities: Record<string, number> } | { type: string, score: number, confidence: number, legend: Record<string, string>, probabilities: Record<string, number> } | { type: string }>, usage: { input_tokens: number, output_tokens: number } }`
1492
1512
 
1493
1513
  ### X
1494
1514
 
@@ -1602,11 +1622,11 @@ Retrieve the captions/transcript for a YouTube video.
1602
1622
  - Output: `{ transcripts: { text: string, start: number }[] }`
1603
1623
 
1604
1624
  #### fetchYoutubeChannel
1605
- Retrieve metadata and recent videos for a YouTube channel.
1625
+ Retrieve the recent videos of a YouTube channel.
1606
1626
  - Accepts a YouTube channel URL (e.g. https://www.youtube.com/@ChannelName or /channel/ID).
1607
- - Returns channel info and video listings as a JSON object.
1627
+ - Returns `videos`, the channel's most recent first, each with id, title, link, views, length, published_time and thumbnail.
1608
1628
  - Input: `{ channelUrl: string }`
1609
- - Output: `object`
1629
+ - Output: `{ videos?: { position?: number, id?: string, title?: string, link?: string, views?: number, length?: string, published_time?: string, channel?: { is_verified?: boolean }, thumbnail?: { static?: string, rich?: string } }[] }`
1610
1630
 
1611
1631
  #### fetchYoutubeComments
1612
1632
  Retrieve comments for a YouTube video.
@@ -1617,10 +1637,11 @@ Retrieve comments for a YouTube video.
1617
1637
 
1618
1638
  #### fetchYoutubeVideo
1619
1639
  Retrieve metadata for a YouTube video (title, description, stats, channel info).
1620
- - Returns video metadata, channel info, and engagement stats.
1640
+ - Returns `video` (title, description, author, length_seconds, views, likes, published_time, thumbnail), `channel` (name, link, subscribers) and `comment.total`, plus chapters, key moments and recommended videos when the video has them.
1641
+ - A field YouTube doesn't show for the video is absent.
1621
1642
  - Video format data is excluded from the response.
1622
1643
  - Input: `{ videoUrl: string }`
1623
- - Output: `object`
1644
+ - Output: `{ video?: { id?: string, title?: string, description?: string, author?: string, category?: string, length_seconds?: number, views?: number, likes?: number, published_time?: string, keywords?: string[], is_family_safe?: boolean, thumbnail?: string }, channel?: { id?: string, name?: string, link?: string, subscribers?: number, thumbnail?: string }, comment?: { total?: number }, chapters?: { title?: string, timestamp?: string, thumbnail?: string }[], key_moments?: { title?: string, start_seconds?: number, thumbnail?: string }[], available_transcripts_languages?: { name?: string, lang?: string }[], recommended_videos?: { videos?: { position?: number, id?: string, title?: string, link?: string, views?: number, channel?: string, length?: string, published_time?: string, thumbnail?: string }[] } }`
1624
1645
 
1625
1646
  #### searchYoutube
1626
1647
  Search for YouTube videos by keyword.
@@ -1633,8 +1654,9 @@ Search for YouTube videos by keyword.
1633
1654
  Retrieve trending videos on YouTube by category and region.
1634
1655
  - Categories: "now" (trending now), "music", "gaming", "films".
1635
1656
  - Supports country and language filtering.
1657
+ - Returns `trending`, each video with id, title, link, views, channel, length, published_time and thumbnail; some categories add `recently_trending` and `shorts`.
1636
1658
  - Input: `{ bp: "now" | "music" | "gaming" | "films", hl: string, gl: string }`
1637
- - Output: `object`
1659
+ - Output: `{ trending?: { position?: number, id?: string, title?: string, link?: string, views?: number, channel?: { id?: string, title?: string, link?: string, is_verified?: boolean, thumbnail?: string }, length?: string, published_time?: string, thumbnail?: string }[], recently_trending?: { position?: number, id?: string, title?: string, link?: string, views?: number, channel?: { id?: string, title?: string, link?: string, is_verified?: boolean, thumbnail?: string }, length?: string, published_time?: string, thumbnail?: string }[], shorts?: { position?: number, id?: string, title?: string, link?: string, views?: number, thumbnail?: string }[] }`
1638
1660
 
1639
1661
  ### Helpers
1640
1662
 
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@mindstudio-ai/agent",
3
- "version": "0.1.181",
3
+ "version": "0.1.182",
4
4
  "description": "TypeScript SDK for MindStudio direct step execution",
5
5
  "type": "module",
6
6
  "main": "./dist/index.js",