@transloadit/node 4.7.6 → 4.7.7

package/src/alphalib/types/robots/_instructions-primitives.ts

@@ -22,12 +22,20 @@ export const robotNames = z.enum([
   'AudioConcatRobot',
   'AudioMergeRobot',
   'AudioArtworkRobot',
+  'AudioSplitRobot',
+  'BoxImportRobot',
+  'BoxStoreRobot',
+  'VideoArtworkRobot',
   'ImageFacedetectRobot',
   'ImageDescribeRobot',
+  'ImageCopyrightdetectRobot',
   'ImageOcrRobot',
   'ImageBgremoveRobot',
+  'ImageUpscaleRobot',
   'ImageGenerateRobot',
+  'ImageEnhanceRobot',
   'DocumentOcrRobot',
+  'DocumentOptimizeRobot',
   'SpeechTranscribeRobot',
   'VideoThumbsRobot',
   'FileVirusscanRobot',
@@ -52,10 +60,12 @@ export const robotNames = z.enum([
   'BackblazeImportRobot',
   'BackblazeStoreRobot',
   'MinioImportRobot',
+  'MegaImportRobot',
   'TigrisImportRobot',
   'CloudflareImportRobot',
   'SupabaseImportRobot',
   'MinioStoreRobot',
+  'MegaStoreRobot',
   'TigrisStoreRobot',
   'CloudflareStoreRobot',
   'SupabaseStoreRobot',
@@ -84,6 +94,8 @@ export const robotNames = z.enum([
   'FileHashRobot',
   'FileReadRobot',
   'VideoOndemandRobot',
+  'VideoSplitRobot',
+  'VideoGenerateRobot',
   'FileFilterRobot',
   'TextSpeakRobot',
   'TextTranslateRobot',
@@ -125,7 +137,6 @@ export const robotMetaSchema = z.object({
   extraChargeForImageResize: z.number().optional(),
 
   // Original keys from content repo:
-  allowed_for_url_transform: z.boolean(),
   bytescount: z.number(),
   description: z.string().optional(),
   discount_factor: z.number(),
@@ -152,6 +163,7 @@ export const robotMetaSchema = z.object({
     'convert',
     'decompress',
     'detect',
+    'enhance',
     'encode',
     'export',
     'extract',
@@ -169,6 +181,7 @@ export const robotMetaSchema = z.object({
     'scan',
     'serve',
     'speak',
+    'split',
     'subtitle',
     'take',
     'transcode',
@@ -181,6 +194,7 @@ export const robotMetaSchema = z.object({
   ]),
   purpose_word: z.string(),
   purpose_words: z.string(),
+  hideCredentialsWarning: z.boolean().optional(),
   requires_credentials: z.literal(true).optional(),
   service_slug: z.enum([
     'artificial-intelligence',
@@ -443,6 +457,8 @@ Allows you to specify a set of metadata that is more expensive on CPU power to c
 
 For images, you can add \`"has_transparency": true\` in this object to extract if the image contains transparent parts and \`"dominant_colors": true\` to extract an array of hexadecimal color codes from the image.
 
+For images, you can also add \`"blurhash": true\` to extract a [BlurHash](https://blurha.sh) string — a compact representation of a placeholder for the image, useful for showing a blurred preview while the full image loads.
+
 For videos, you can add the \`"colorspace: true"\` parameter to extract the colorspace of the output video.
 
 For audio, you can add \`"mean_volume": true\` to get a single value representing the mean average volume of the audio file.
@@ -571,6 +587,22 @@ Specifies which Step(s) to use as input.
     ]
   }
   \`\`\`
+- You can also tag an input Step with \`as\` to pass semantic intent to robots:
+- You can also tag input Steps with \`as\` to pass semantic intent to robots:
+  \`\`\`json
+  {
+    "use": [
+      {
+        "name": ":original",
+        "as": "image"
+      },
+      {
+        "name": ":original",
+        "as": "mask"
+      }
+    ]
+  }
+  \`\`\`
 
 > [!Tip]
 > That's likely all you need to know about \`use\`, but you can view [Advanced use cases](/docs/topics/use-parameter/).
@@ -600,6 +632,22 @@ Specifies which Step(s) to use as input.
     ]
   }
   \`\`\`
+- You can also tag an input Step with \`as\` to pass semantic intent to robots:
+- You can also tag input Steps with \`as\` to pass semantic intent to robots:
+  \`\`\`json
+  {
+    "use": [
+      {
+        "name": ":original",
+        "as": "image"
+      },
+      {
+        "name": ":original",
+        "as": "mask"
+      }
+    ]
+  }
+  \`\`\`
 
 > [!Tip]
 > That's likely all you need to know about \`use\`, but you can view [Advanced use cases](/docs/topics/use-parameter/).
@@ -621,7 +669,7 @@ export const complexWidthSchema = z.preprocess((val) => {
     return num
   }
   return val
-}, z.number().int().min(1).max(7680))
+}, z.number().int().min(1).max(25_000))
 
 export const complexHeightSchema = z.preprocess((val) => {
   if (typeof val === 'string' && val.startsWith('${')) {
@@ -635,7 +683,7 @@ export const complexHeightSchema = z.preprocess((val) => {
     return num
   }
   return val
-}, z.number().int().min(1).max(4320))
+}, z.number().int().min(1).max(25_000))
 
 /**
  * A robot that uses FFmpeg.
@@ -1252,6 +1300,24 @@ While we recommend to use <dfn>Template Credentials</dfn> at all times, some use
   })
   .strict()
 
+export type BoxBase = z.infer<typeof boxBase>
+export const boxBase = z
+  .object({
+    credentials: z
+      .string()
+      .optional()
+      .describe(`
+Create a Box app in the [Box Developer Console](https://app.box.com/developers/console) using **Server Authentication (JWT)**.
+
+Generate a public/private keypair for that app. Box will download a JSON config file (with \`boxAppSettings\` and \`appAuth\`) that includes your JWT private key material.
+
+In your Transloadit account, create [Template Credentials](/c/template-credentials/) of type Box and paste that full JSON file into \`key_file_contents\`. Then set this \`credentials\` parameter to the Template Credentials name.
+
+If your Box enterprise requires it, make sure the app is authorized by an admin before using it in production.
+`),
+  })
+  .strict()
+
 export type CloudfilesBase = z.infer<typeof cloudfilesBase>
 export const cloudfilesBase = z
   .object({
@@ -1389,6 +1455,30 @@ While we recommend to use <dfn>Template Credentials</dfn> at all times, some use
   })
   .strict()
 
+export type MegaBase = z.infer<typeof megaBase>
+export const megaBase = z
+  .object({
+    credentials: z
+      .string()
+      .optional()
+      .describe(`
+Please create your associated <dfn>Template Credentials</dfn> in your Transloadit account and use the name of your <dfn>Template Credentials</dfn> as this parameter's value. They will contain the values for your MEGA object storage bucket, Key, Secret and Bucket region.
+
+While we recommend to use <dfn>Template Credentials</dfn> at all times, some use cases demand dynamic credentials for which using <dfn>Template Credentials</dfn> is too unwieldy because of their static nature. If you have this requirement, feel free to use the following parameters instead: \`"bucket"\`, \`"bucket_region"\`, \`"key"\`, \`"secret"\`.
+`),
+    bucket: z.string().optional(),
+    bucket_region: z
+      .string()
+      .optional()
+      .describe(`
+The region where the bucket is located.
+`),
+    host: z.string().optional(),
+    key: z.string().optional(),
+    secret: z.string().optional(),
+  })
+  .strict()
+
 export type S3Base = z.infer<typeof s3Base>
 export const s3Base = z
   .object({

package/src/alphalib/types/robots/ai-chat.ts

@@ -115,10 +115,20 @@ const coreMessageSchema = z.discriminatedUnion('role', [
 
 export const meta: RobotMetaInput = {
   name: 'AiChatRobot',
-  allowed_for_url_transform: true,
   bytescount: 1,
   discount_factor: 1,
   discount_pct: 0,
+  example_code: {
+    steps: {
+      reply: {
+        robot: '/ai/chat',
+        model: 'auto',
+        messages:
+          'Summarize this in one sentence: Transloadit handles uploads and media processing.',
+      },
+    },
+  },
+  example_code_description: 'Generate a concise AI response from a text prompt:',
   minimum_charge: 0,
   output_factor: 0.6,
   purpose_sentence: 'generates AI chat responses from prompts',
@@ -132,7 +142,7 @@ export const meta: RobotMetaInput = {
   typical_file_type: 'document',
   priceFactor: 1,
   queueSlotCount: 10,
-  // Is this a sensbile minimum charge? What if the customer supplies their own keys? Is it low enough for these cases?
+  // Is this a sensible minimum charge? What if the customer supplies their own keys? Is it low enough for these cases?
   minimumChargeUsd: 0.06,
   isAllowedForUrlTransform: true,
   trackOutputFileSize: true,
@@ -148,7 +158,6 @@ export const meta: RobotMetaInput = {
 export const MODEL_CAPABILITIES: Record<string, { pdf: boolean; image: boolean }> = {
   'anthropic/claude-4-sonnet-20250514': { pdf: true, image: true },
   'anthropic/claude-4-opus-20250514': { pdf: true, image: true },
-  'anthropic/claude-sonnet-4-6': { pdf: true, image: true },
   'anthropic/claude-sonnet-4-5': { pdf: true, image: true },
   'anthropic/claude-opus-4-5': { pdf: true, image: true },
   'anthropic/claude-opus-4-6': { pdf: true, image: true },
@@ -160,10 +169,19 @@ export const MODEL_CAPABILITIES: Record<string, { pdf: boolean; image: boolean }
   'openai/gpt-5.2-2025-12-11': { pdf: false, image: true },
   'openai/gpt-5.2-chat-latest': { pdf: false, image: true },
   'openai/gpt-5.2-pro': { pdf: false, image: true },
+  'openai/gpt-5.4': { pdf: false, image: true },
+  'openai/gpt-5.4-mini': { pdf: false, image: true },
+  'openai/gpt-5.4-nano': { pdf: false, image: true },
   'google/gemini-2.5-pro': { pdf: true, image: true },
   'moonshot/kimi-k2': { pdf: false, image: false },
 }
 
+// Default model for /ai/chat when `model: "auto"` (or unset).
+// 2026-02-10: default is Opus 4.6 (intentional; aligns with our current recommended Anthropic
+// flagship model and the system tests in this repo). Keep this aligned with MODEL_CAPABILITIES.
+export const AI_CHAT_DEFAULT_MODEL =
+  'anthropic/claude-opus-4-6' satisfies keyof typeof MODEL_CAPABILITIES
+
 const supportedModelsList = Object.keys(MODEL_CAPABILITIES)
 
 export const vendorModelSchema = z
@@ -179,7 +197,7 @@ export const robotAiChatInstructionsSchema = robotBase
   .merge(robotUse)
   .extend({
     robot: z.literal('/ai/chat'),
-    // TODO: Is the auto mode yet implemented?
+    // NOTE: model:"auto" is resolved server-side to AI_CHAT_DEFAULT_MODEL for now.
     model: z
       .union([vendorModelSchema, z.literal('auto')])
       .default('auto')
@@ -196,6 +214,12 @@ export const robotAiChatInstructionsSchema = robotBase
       .string()
       .optional()
       .describe('Set the system/developer prompt, if the model allows it'),
+    reasoning_effort: z
+      .enum(['high', 'medium', 'low'])
+      .optional()
+      .describe(
+        'Controls how much effort the model spends on reasoning. Higher values produce more thorough responses but cost more tokens. Applies to models that support extended thinking (OpenAI o-series, GPT-5.x, Anthropic Claude with thinking). If omitted, the model default is used.',
+      ),
     credentials: z
       .union([z.string(), z.array(z.string())])
       .optional()
@@ -215,11 +239,17 @@ export const robotAiChatInstructionsSchema = robotBase
           url: z.string(),
           headers: z.record(z.string()).optional(),
           auth: z.enum(['transloadit']).optional(),
+          allowed_tools: z
+            .array(z.string())
+            .optional()
+            .describe(
+              'Optional allowlist of tool names to expose from this MCP server. If omitted, all tools exposed by the server are available to the model.',
+            ),
         }),
       )
       .optional()
       .describe(
-        'The MCP servers to use. This is used to call tools from the LLM. Use `headers` to pass `Authorization: Bearer <token>` when needed. You can use any MCP server reachable from your environment. For Transloadit\'s own MCP server, you can set `auth: "transloadit"` to let API2 auto-auth and inject an Authorization header for you (only for Transloadit-hosted MCP servers).',
+        'The MCP servers to use for tool calling. You can use any MCP server reachable from your environment. Use `headers` to pass server-specific auth (for example `Authorization: Bearer <token>`). For Transloadit\'s MCP server: Bearer tokens minted via `/token` satisfy Signature Authentication (signature checks apply only to key/secret requests). `auth: "transloadit"` is reserved for API2-managed auth to Transloadit-hosted MCP servers.',
       ),
   })
   .strict()

package/src/alphalib/types/robots/assembly-savejson.ts

@@ -5,6 +5,14 @@ import { interpolateRobot, robotBase } from './_instructions-primitives.ts'
 // @ts-expect-error - AssemblySavejsonRobot is not ready yet @TODO please supply missing keys
 export const meta: RobotMetaInput = {
   name: 'AssemblySavejsonRobot',
+  example_code: {
+    steps: {
+      save_json: {
+        robot: '/assembly/savejson',
+      },
+    },
+  },
+  example_code_description: 'Save Assembly result data as JSON:',
   priceFactor: 0,
   queueSlotCount: 5,
   isAllowedForUrlTransform: true,

package/src/alphalib/types/robots/audio-artwork.ts

@@ -10,7 +10,6 @@ import {
 } from './_instructions-primitives.ts'
 
 export const meta: RobotMetaInput = {
-  allowed_for_url_transform: true,
   bytescount: 1,
   discount_factor: 1,
   discount_pct: 0,
@@ -28,7 +27,7 @@ export const meta: RobotMetaInput = {
   output_factor: 0.8,
   override_lvl1: 'Audio Encoding',
   purpose_sentence:
-    'extracts the embedded cover artwork from audio files and allows you to pipe it into other Steps, for example into /image/resize Steps. It can also insert images into audio files as cover artwork',
+    'extracts embedded cover artwork from audio files or inserts a new cover image into them. Extracted artwork can be piped into other Steps such as /image/resize. Use `method: "insert"` to embed artwork into audio files like MP3, FLAC, or M4A',
   purpose_verb: 'extract',
   purpose_word: 'extract/insert artwork',
   purpose_words: 'Extract or insert audio artwork',
@@ -53,11 +52,11 @@ export const robotAudioArtworkInstructionsSchema = robotBase
   .merge(robotFFmpegAudio)
   .extend({
     robot: z.literal('/audio/artwork').describe(`
-For extraction, this <dfn>Robot</dfn> uses the image format embedded within the audio file — most often, this is JPEG.
+This <dfn>Robot</dfn> extracts or inserts cover artwork in audio files.
 
-If you need the image in a different format, pipe the result of this <dfn>Robot</dfn> into [🤖/image/resize](/docs/robots/image-resize/).
+For extraction, it uses the image format embedded within the audio file — most often, this is JPEG. If you need the image in a different format, pipe the result into [🤖/image/resize](/docs/robots/image-resize/).
 
-The \`method\` parameter determines whether to extract or insert.
+For insertion, provide both an audio file (as \`"audio"\`) and an image file (as \`"image"\`) via the \`use\` parameter, and set \`method\` to \`"insert"\`. The image will be embedded as the cover artwork of the audio file.
 `),
     method: z
       .enum(['extract', 'insert'])

package/src/alphalib/types/robots/audio-concat.ts

@@ -13,7 +13,6 @@ import {
 } from './_instructions-primitives.ts'
 
 export const meta: RobotMetaInput = {
-  allowed_for_url_transform: false,
   bytescount: 4,
   discount_factor: 0.25,
   discount_pct: 75,
@@ -62,7 +61,7 @@ export const meta: RobotMetaInput = {
   name: 'AudioConcatRobot',
   priceFactor: 4,
   queueSlotCount: 20,
-  isAllowedForUrlTransform: false,
+  isAllowedForUrlTransform: true,
   trackOutputFileSize: true,
   isInternal: false,
   removeJobResultFilesFromDiskRightAfterStoringOnS3: false,

package/src/alphalib/types/robots/audio-encode.ts

@@ -12,7 +12,6 @@ import {
 } from './_instructions-primitives.ts'
 
 export const meta: RobotMetaInput = {
-  allowed_for_url_transform: false,
   bytescount: 4,
   discount_factor: 0.25,
   discount_pct: 75,
@@ -45,7 +44,7 @@ export const meta: RobotMetaInput = {
   name: 'AudioEncodeRobot',
   priceFactor: 4,
   queueSlotCount: 20,
-  isAllowedForUrlTransform: false,
+  isAllowedForUrlTransform: true,
   trackOutputFileSize: true,
   isInternal: false,
   stage: 'ga',

package/src/alphalib/types/robots/audio-loop.ts

@@ -12,7 +12,6 @@ import {
 } from './_instructions-primitives.ts'
 
 export const meta: RobotMetaInput = {
-  allowed_for_url_transform: false,
   bytescount: 4,
   discount_factor: 0.25,
   discount_pct: 75,
@@ -45,7 +44,7 @@ export const meta: RobotMetaInput = {
   name: 'AudioLoopRobot',
   priceFactor: 4,
   queueSlotCount: 20,
-  isAllowedForUrlTransform: false,
+  isAllowedForUrlTransform: true,
   trackOutputFileSize: true,
   isInternal: false,
   removeJobResultFilesFromDiskRightAfterStoringOnS3: false,

package/src/alphalib/types/robots/audio-merge.ts

@@ -13,7 +13,6 @@ import {
 } from './_instructions-primitives.ts'
 
 export const meta: RobotMetaInput = {
-  allowed_for_url_transform: false,
   bytescount: 4,
   discount_factor: 0.25,
   discount_pct: 75,
@@ -63,7 +62,7 @@ export const meta: RobotMetaInput = {
   name: 'AudioMergeRobot',
   priceFactor: 4,
   queueSlotCount: 20,
-  isAllowedForUrlTransform: false,
+  isAllowedForUrlTransform: true,
   trackOutputFileSize: true,
   isInternal: false,
   removeJobResultFilesFromDiskRightAfterStoringOnS3: false,

package/src/alphalib/types/robots/audio-split.ts

@@ -0,0 +1,130 @@
+import { z } from 'zod'
+
+import { stackVersions } from '../stackVersions.ts'
+import type { RobotMetaInput } from './_instructions-primitives.ts'
+import {
+  bitrateSchema,
+  interpolateRobot,
+  robotBase,
+  robotFFmpegAudio,
+  robotUse,
+  sampleRateSchema,
+} from './_instructions-primitives.ts'
+
+export const meta: RobotMetaInput = {
+  bytescount: 4,
+  discount_factor: 0.25,
+  discount_pct: 75,
+  example_code: {
+    steps: {
+      split: {
+        robot: '/audio/split',
+        use: ':original',
+        ffmpeg_stack: stackVersions.ffmpeg.recommendedVersion,
+        segments: [
+          { from: 0, to: 30 },
+          { from: 60, to: 90 },
+        ],
+      },
+    },
+  },
+  example_code_description:
+    'Split an audio file into two segments, extracting the first 30 seconds and a segment from 1:00 to 1:30:',
+  minimum_charge: 0,
+  output_factor: 0.8,
+  override_lvl1: 'Audio Encoding',
+  purpose_sentence:
+    'splits an audio file into multiple segments based on an array of from/to durations',
+  purpose_verb: 'split',
+  purpose_word: 'split',
+  purpose_words: 'Split audio',
+  service_slug: 'audio-encoding',
+  slot_count: 20,
+  title: 'Split audio',
+  typical_file_size_mb: 3.8,
+  typical_file_type: 'audio file',
+  uses_tools: ['ffmpeg'],
+  name: 'AudioSplitRobot',
+  priceFactor: 4,
+  queueSlotCount: 20,
+  isAllowedForUrlTransform: true,
+  trackOutputFileSize: true,
+  isInternal: false,
+  removeJobResultFilesFromDiskRightAfterStoringOnS3: false,
+  stage: 'ga',
+}
+
+const segmentSchema = z
+  .object({
+    from: z
+      .union([z.number(), z.string()])
+      .describe(
+        'Start time of the segment in seconds (e.g. `30`) or as a timecode string (e.g. `"00:00:30.000"`).',
+      ),
+    to: z
+      .union([z.number(), z.string()])
+      .describe(
+        'End time of the segment in seconds (e.g. `60`) or as a timecode string (e.g. `"00:01:00.000"`).',
+      ),
+  })
+  .strict()
+
+export const robotAudioSplitInstructionsSchema = robotBase
+  .merge(robotUse)
+  .merge(robotFFmpegAudio)
+  .extend({
+    result: z
+      .boolean()
+      .optional()
+      .describe('Whether the results of this Step should be present in the Assembly Status JSON'),
+    robot: z.literal('/audio/split').describe(`
+Splits an audio file into multiple segments based on an array of from/to durations. Each segment produces a separate output file. This is useful for cutting an audio file into parts, for example to insert ads later via [🤖/audio/concat](/docs/robots/audio-concat/).
+`),
+    segments: z
+      .array(segmentSchema)
+      .min(1)
+      .describe(`
+An array of objects, each specifying a segment to extract from the input audio. Each object must have a \`from\` and \`to\` key indicating the start and end time of the segment.
+
+Times can be specified as numbers (seconds) or as timecode strings (e.g. \`"00:01:30.000"\`).
+`),
+    bitrate: bitrateSchema.optional().describe(`
+Bit rate of the resulting audio file, in bits per second. If not specified will default to the bit rate of the input audio file.
+`),
+    sample_rate: sampleRateSchema.optional().describe(`
+Sample rate of the resulting audio file, in Hertz. If not specified will default to the sample rate of the input audio file.
+`),
+  })
+  .strict()
+
+export const robotAudioSplitInstructionsWithHiddenFieldsSchema =
+  robotAudioSplitInstructionsSchema.extend({
+    result: z
+      .union([z.literal('debug'), robotAudioSplitInstructionsSchema.shape.result])
+      .optional(),
+  })
+
+export type RobotAudioSplitInstructions = z.infer<typeof robotAudioSplitInstructionsSchema>
+export type RobotAudioSplitInstructionsWithHiddenFields = z.infer<
+  typeof robotAudioSplitInstructionsWithHiddenFieldsSchema
+>
+
+export const interpolatableRobotAudioSplitInstructionsSchema = interpolateRobot(
+  robotAudioSplitInstructionsSchema,
+)
+export type InterpolatableRobotAudioSplitInstructions =
+  InterpolatableRobotAudioSplitInstructionsInput
+
+export type InterpolatableRobotAudioSplitInstructionsInput = z.input<
+  typeof interpolatableRobotAudioSplitInstructionsSchema
+>
+
+export const interpolatableRobotAudioSplitInstructionsWithHiddenFieldsSchema = interpolateRobot(
+  robotAudioSplitInstructionsWithHiddenFieldsSchema,
+)
+export type InterpolatableRobotAudioSplitInstructionsWithHiddenFields = z.infer<
+  typeof interpolatableRobotAudioSplitInstructionsWithHiddenFieldsSchema
+>
+export type InterpolatableRobotAudioSplitInstructionsWithHiddenFieldsInput = z.input<
+  typeof interpolatableRobotAudioSplitInstructionsWithHiddenFieldsSchema
+>