HiruAI Creative Cloud is an MCP server published by com.hiruai, hosted at mcp.hiruai.com. When last checked on 2026-11-22 it was working: it answers and lists its tools. It declares 34 tools, including image_create, edit_create, upscale_create, voice_create, music_create.
Generate images, voice, music and video, then edit them into a finished movie from one plan.
Nothing else in the archive shares these values.
Squares are this server's values; dots are other servers sharing them. A shared address or publisher is a lead, not proof of a common operator.
Inferred from the host name and headers: what a server runs on, not who runs it.
Submit image work for the authenticated account. Returns an asynchronous request ID. For a sequence of work, prefer batch_submit and hand off the dashboard URL without polling. Use library IDs from assets_list where supported. Rights/privacy acknowledgements must reflect the user’s authorization; never invent consent. Limits remain active through result email delivery. Creative baseline: unless the user declined, build at least one dopamine hook from dopamine_hooks_get into this prompt, script or sound. First call image_models_get. Present dated original-model Arena scores (not scores of our quantized/LoRA recipes), traits, availability and native alternatives to help the user choose. Select one model; never invent agreement or substitute user consent for the operator's hosting permission. For quality-first image generation or editing, prefer the connected agent’s own image tool when it supports the requested operation, for example ChatGPT image generation and image editing. Check what this client actually exposes, including in Claude. Honor the user’s chosen provider. For anime-style character illustration, HiruAI Image offers WAI Illustrious SDXL v17 when available in image_models_get.
| Parameter | Type | Description |
|---|---|---|
| model |
| string |
| One model per new request; Qwen Image 2512 Q4 + Lightning (8-step) is the default. The separate 50-step quality model remains selectable. Read image_models_get first and present dated quality evidence, strengths, availability and native alternatives; no Lightning Arena score is measured. Creator-res… |
| prompt * | string | Image prompt, 3–2000 characters. Whitespace is collapsed before backend length/safety checks. |
| nsfw_requested | boolean | Legacy alias selecting only Z-Image Turbo NSFW. Prefer model; never request two models together. |
| idempotency_key * | string | Unique per intended operation. Reuse on network retries. |
| acknowledge_adult | boolean | Confirm that the requesting user meets the adult-use requirement. |
| acknowledge_rights * | boolean | Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent. |
| acknowledge_privacy * | boolean | Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent. |
| model_terms_version | string | Current catalog version from image_models_get, required with creator model terms. |
| acknowledge_model_terms | boolean | Use true only after actual user authorization for the selected model. User consent never grants the platform hosting rights. |
| acknowledge_nsfw_policy | boolean | Confirm the user accepts the selected model’s adult-content restrictions. |
| acknowledge_noncommercial | boolean | Use true only after actual user authorization for the selected model. User consent never grants the platform hosting rights. |
| acknowledge_no_real_person | boolean | Confirm the prompt and inputs do not depict or imitate a real person for adult content. |
{
"name": "image_create",
"title": "Create image",
"description": "Submit image work for the authenticated account. Returns an asynchronous request ID. For a sequence of work, prefer batch_submit and hand off the dashboard URL without polling. Use library IDs from assets_list where supported. Rights/privacy acknowledgements must reflect the user’s authorization; never invent consent. Limits remain active through result email delivery. Creative baseline: unless the user declined, build at least one dopamine hook from dopamine_hooks_get into this prompt, script or sound. First call image_models_get. Present dated original-model Arena scores (not scores of our quantized/LoRA recipes), traits, availability and native alternatives to help the user choose. Select one model; never invent agreement or substitute user consent for the operator's hosting permission. For quality-first image generation or editing, prefer the connected agent’s own image tool when it supports the requested operation, for example ChatGPT image generation and image editing. Check what this client actually exposes, including in Claude. Honor the user’s chosen provider. For anime-style character illustration, HiruAI Image offers WAI Illustrious SDXL v17 when available in image_models_get.",
"inputSchema": {
"type": "object",
"allOf": [
{
"if": {
"anyOf": [
{
"required": [
"nsfw_requested"
],
"properties": {
"nsfw_requested": {
"const": true
}
}
},
{
"required": [
"model"
],
"properties": {
"model": {
"const": "zimage-turbo-q4k-nsfw-master-lora-pascal"
}
}
}
]
},
"then": {
"required": [
"acknowledge_adult",
"acknowledge_nsfw_policy",
"acknowledge_no_real_person"
],
"properties": {
"acknowledge_adult": {
"const": true
},
"acknowledge_nsfw_policy": {
"const": true
},
"acknowledge_no_real_person": {
"const": true
}
}
}
},
{
"if": {
"required": [
"model"
],
"properties": {
"model": {
"enum": [
"wai-illustrious-sdxl-v17",
"pony-realism-v2.3",
"prefect-pony-xl-v6",
"pony-diffusion-v6-xl"
]
}
}
},
"then": {
"required": [
"acknowledge_model_terms",
"model_terms_version"
],
"properties": {
"model_terms_version": {
"minLength": 1
},
"acknowledge_model_terms": {
"const": true
}
}
}
},
{
"if": {
"required": [
"model"
],
"properties": {
"model": {
"const": "prefect-pony-xl-v6"
}
}
},
"then": {
"required": [
"acknowledge_noncommercial"
],
"properties": {
"acknowledge_noncommercial": {
"const": true
}
}
}
}
],
"required": [
"prompt",
"acknowledge_rights",
"acknowledge_privacy",
"idempotency_key"
],
"properties": {
"model": {
"enum": [
"qwen-image-2512-q4-lightning",
"qwen-image-2512-q4-quality",
"zimage-turbo-q4k-pascal",
"zimage-turbo-q4k-nsfw-master-lora-pascal",
"wai-illustrious-sdxl-v17",
"pony-realism-v2.3",
"prefect-pony-xl-v6",
"pony-diffusion-v6-xl"
],
"type": "string",
"default": "qwen-image-2512-q4-lightning",
"maxLength": 100,
"description": "One model per new request; Qwen Image 2512 Q4 + Lightning (8-step) is the default. The separate 50-step quality model remains selectable. Read image_models_get first and present dated quality evidence, strengths, availability and native alternatives; no Lightning Arena score is measured. Creator-restricted choices require recorded hosting permission."
},
"prompt": {
"type": "string",
"examples": [
"A playful orange astronaut cat DJ on the Moon, cinematic mint lighting."
],
"maxLength": 2000,
"minLength": 3,
"description": "Image prompt, 3–2000 characters. Whitespace is collapsed before backend length/safety checks."
},
"nsfw_requested": {
"type": "boolean",
"default": false,
"description": "Legacy alias selecting only Z-Image Turbo NSFW. Prefer model; never request two models together."
},
"idempotency_key": {
"type": "string",
"maxLength": 128,
"minLength": 8,
"description": "Unique per intended operation. Reuse on network retries."
},
"acknowledge_adult": {
"type": "boolean",
"default": false,
"description": "Confirm that the requesting user meets the adult-use requirement."
},
"acknowledge_rights": {
"type": "boolean",
"const": true,
"description": "Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent."
},
"acknowledge_privacy": {
"type": "boolean",
"const": true,
"description": "Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent."
},
"model_terms_version": {
"type": "string",
"maxLength": 80,
"description": "Current catalog version from image_models_get, required with creator model terms."
},
"acknowledge_model_terms": {
"type": "boolean",
"default": false,
"description": "Use true only after actual user authorization for the selected model. User consent never grants the platform hosting rights."
},
"acknowledge_nsfw_policy": {
"type": "boolean",
"default": false,
"description": "Confirm the user accepts the selected model’s adult-content restrictions."
},
"acknowledge_noncommercial": {
"type": "boolean",
"default": false,
"description": "Use true only after actual user authorization for the selected model. User consent never grants the platform hosting rights."
},
"acknowledge_no_real_person": {
"type": "boolean",
"default": false,
"description": "Confirm the prompt and inputs do not depict or imitate a real person for adult content."
}
},
"description": "Create one 1024x1024 image from one selected model. Read image_models_get and explain quality evidence and applicable consent before selection. Existing older paired results remain readable. Prompt 3–2000 characters.",
"additionalProperties": false
},
"annotations": {
"title": "Create image",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Submit edit work for the authenticated account. Returns an asynchronous request ID. For a sequence of work, prefer batch_submit and hand off the dashboard URL without polling. Use library IDs from assets_list where supported. Rights/privacy acknowledgements must reflect the user’s authorization; never invent consent. Limits remain active through result email delivery. For quality-first image generation or editing, prefer the connected agent’s own image tool when it supports the requested operation, for example ChatGPT image generation and image editing. Check what this client actually exposes, including in Claude. Honor the user’s chosen provider. For anime-style character illustration, HiruAI Image offers WAI Illustrious SDXL v17 when available in image_models_get.
| Parameter | Type | Description |
|---|---|---|
| files | array | Optional upload parts (max 20 MiB total). Use audio for Voice, file for Edit/Upscale, image_upload/last_image_upload/voice_upload/music_upload for Video. |
| uploads | array | Private uploaded file references from asset_upload or the web file library. No local paths, URLs or base64 in plans. Set the corresponding source/image/voice/music mode to upload. |
| instruction * | string | Editing instruction, 3–2000 characters after whitespace normalization. |
| source_type * | string | Choose upload for file inputs or generated for owned HiruAI results. Do not mix both kinds. |
| idempotency_key * | string | Unique per intended operation. Reuse on network retries. |
| generated_service | array | One service per selected image; omission defaults every image to image. When supplied, length must match generated_request_id and generated_variant. |
| generated_variant | array | Variants aligned by index: image base/nsfw; edit result; upscale upscaled/4k. |
| acknowledge_rights * | boolean | Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent. |
| acknowledge_safety * | boolean | Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent. |
| acknowledge_privacy * | boolean | Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent. |
| generated_request_id | array | 1–3 owned completed request UUIDs, aligned by index with services and variants. Use bindings for future step outputs. |
{
"name": "edit_create",
"title": "Edit image",
"description": "Submit edit work for the authenticated account. Returns an asynchronous request ID. For a sequence of work, prefer batch_submit and hand off the dashboard URL without polling. Use library IDs from assets_list where supported. Rights/privacy acknowledgements must reflect the user’s authorization; never invent consent. Limits remain active through result email delivery. For quality-first image generation or editing, prefer the connected agent’s own image tool when it supports the requested operation, for example ChatGPT image generation and image editing. Check what this client actually exposes, including in Claude. Honor the user’s chosen provider. For anime-style character illustration, HiruAI Image offers WAI Illustrious SDXL v17 when available in image_models_get.",
"inputSchema": {
"type": "object",
"allOf": [
{
"if": {
"required": [
"source_type"
],
"properties": {
"source_type": {
"const": "generated"
}
}
},
"then": {
"required": [
"generated_request_id",
"generated_variant"
],
"properties": {
"generated_variant": {
"minItems": 1
},
"generated_request_id": {
"minItems": 1
}
}
}
},
{
"if": {
"required": [
"source_type"
],
"properties": {
"source_type": {
"const": "upload"
}
}
},
"then": {
"properties": {
"generated_service": {
"maxItems": 0
},
"generated_variant": {
"maxItems": 0
},
"generated_request_id": {
"maxItems": 0
}
}
}
}
],
"required": [
"source_type",
"instruction",
"acknowledge_rights",
"acknowledge_privacy",
"acknowledge_safety",
"idempotency_key"
],
"properties": {
"files": {
"type": "array",
"items": {
"type": "object",
"required": [
"field",
"filename",
"mime_type",
"base64"
],
"properties": {
"field": {
"type": "string"
},
"base64": {
"type": "string"
},
"filename": {
"type": "string"
},
"mime_type": {
"type": "string"
}
},
"additionalProperties": false
},
"maxItems": 3,
"description": "Optional upload parts (max 20 MiB total). Use audio for Voice, file for Edit/Upscale, image_upload/last_image_upload/voice_upload/music_upload for Video."
},
"uploads": {
"type": "array",
"items": {
"type": "object",
"title": "UploadInput",
"required": [
"field",
"asset_id"
],
"properties": {
"field": {
"type": "string",
"title": "Field"
},
"asset_id": {
"type": "string",
"title": "Asset Id",
"format": "uuid"
}
},
"additionalProperties": false
},
"maxItems": 3,
"minItems": 1,
"description": "Private uploaded file references from asset_upload or the web file library. No local paths, URLs or base64 in plans. Set the corresponding source/image/voice/music mode to upload."
},
"instruction": {
"type": "string",
"maxLength": 2000,
"minLength": 3,
"description": "Editing instruction, 3–2000 characters after whitespace normalization."
},
"source_type": {
"enum": [
"upload",
"generated"
],
"type": "string",
"description": "Choose upload for file inputs or generated for owned HiruAI results. Do not mix both kinds."
},
"idempotency_key": {
"type": "string",
"maxLength": 128,
"minLength": 8,
"description": "Unique per intended operation. Reuse on network retries."
},
"generated_service": {
"type": "array",
"items": {
"enum": [
"image",
"edit",
"upscale"
],
"type": "string"
},
"default": [],
"maxItems": 3,
"description": "One service per selected image; omission defaults every image to image. When supplied, length must match generated_request_id and generated_variant."
},
"generated_variant": {
"type": "array",
"items": {
"enum": [
"base",
"nsfw",
"result",
"upscaled",
"4k"
],
"type": "string"
},
"default": [],
"maxItems": 3,
"description": "Variants aligned by index: image base/nsfw; edit result; upscale upscaled/4k."
},
"acknowledge_rights": {
"type": "boolean",
"const": true,
"description": "Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent."
},
"acknowledge_safety": {
"type": "boolean",
"const": true,
"description": "Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent."
},
"acknowledge_privacy": {
"type": "boolean",
"const": true,
"description": "Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent."
},
"generated_request_id": {
"type": "array",
"items": {
"type": "string",
"format": "uuid"
},
"default": [],
"maxItems": 3,
"description": "1–3 owned completed request UUIDs, aligned by index with services and variants. Use bindings for future step outputs."
}
},
"description": "Edit 1–3 images, all from uploads or all from owned generated results. Parallel generated_* arrays must have equal lengths and unique (service,id,variant) triples. Image supports base/nsfw, Edit result, Upscale upscaled/4k.",
"additionalProperties": false
},
"annotations": {
"title": "Edit image",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Submit upscale work for the authenticated account. Returns an asynchronous request ID. For a sequence of work, prefer batch_submit and hand off the dashboard URL without polling. Use library IDs from assets_list where supported. Rights/privacy acknowledgements must reflect the user’s authorization; never invent consent. Limits remain active through result email delivery.
| Parameter | Type | Description |
|---|---|---|
| files | array | Optional upload parts (max 20 MiB total). Use audio for Voice, file for Edit/Upscale, image_upload/last_image_upload/voice_upload/music_upload for Video. |
| uploads | array | Private uploaded file references from asset_upload or the web file library. No local paths, URLs or base64 in plans. Set the corresponding source/image/voice/music mode to upload. |
| source_url | ["string","null"] | For source_type=url only: a public HTTPS image URL, max 2048 characters; no credentials, IP literals, non-443 ports or private network destinations. |
| source_type * | string | Choose an uploaded source or a previous generated image. |
| idempotency_key * | string | Unique per intended operation. Reuse on network retries. |
| generated_service | ["string","null"] | For generated input: image or edit. Omission defaults to image. |
| generated_variant | ["string","null"] | For generated input: image uses base/nsfw; edit uses result. |
| acknowledge_rights * | boolean | Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent. |
| acknowledge_privacy * | boolean | Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent. |
| generated_request_id | ["string","null"] | Owned completed source request UUID; required with source_type=generated. |
{
"name": "upscale_create",
"title": "Upscale image",
"description": "Submit upscale work for the authenticated account. Returns an asynchronous request ID. For a sequence of work, prefer batch_submit and hand off the dashboard URL without polling. Use library IDs from assets_list where supported. Rights/privacy acknowledgements must reflect the user’s authorization; never invent consent. Limits remain active through result email delivery.",
"inputSchema": {
"type": "object",
"allOf": [
{
"if": {
"required": [
"source_type"
],
"properties": {
"source_type": {
"const": "url"
}
}
},
"then": {
"required": [
"source_url"
],
"properties": {
"source_url": {
"type": "string",
"minLength": 1
}
}
}
},
{
"if": {
"required": [
"source_type"
],
"properties": {
"source_type": {
"const": "generated"
}
}
},
"then": {
"required": [
"generated_request_id",
"generated_variant"
],
"properties": {
"generated_variant": {
"type": "string"
},
"generated_request_id": {
"type": "string"
}
}
}
},
{
"if": {
"required": [
"source_type"
],
"properties": {
"source_type": {
"const": "upload"
}
}
},
"then": {
"properties": {
"generated_service": {
"enum": [
null,
""
]
},
"generated_variant": {
"enum": [
null,
""
]
},
"generated_request_id": {
"enum": [
null,
""
]
}
}
}
},
{
"if": {
"required": [
"source_type"
],
"properties": {
"source_type": {
"const": "url"
}
}
},
"then": {
"properties": {
"generated_service": {
"enum": [
null,
""
]
},
"generated_variant": {
"enum": [
null,
""
]
},
"generated_request_id": {
"enum": [
null,
""
]
}
}
}
},
{
"if": {
"required": [
"source_type"
],
"properties": {
"source_type": {
"const": "upload"
}
}
},
"then": {
"properties": {
"source_url": {
"enum": [
null,
""
]
}
}
}
},
{
"if": {
"required": [
"source_type"
],
"properties": {
"source_type": {
"const": "generated"
}
}
},
"then": {
"properties": {
"source_url": {
"enum": [
null,
""
]
}
}
}
},
{
"if": {
"required": [
"generated_service"
],
"properties": {
"generated_service": {
"const": "image"
}
}
},
"then": {
"properties": {
"generated_variant": {
"enum": [
"base",
"nsfw"
]
}
}
}
},
{
"if": {
"required": [
"generated_service"
],
"properties": {
"generated_service": {
"const": "edit"
}
}
},
"then": {
"properties": {
"generated_variant": {
"enum": [
"result"
]
}
}
}
},
{
"if": {
"not": {
"required": [
"generated_service"
]
}
},
"then": {
"properties": {
"generated_variant": {
"enum": [
"base",
"nsfw",
null
]
}
}
}
}
],
"required": [
"source_type",
"acknowledge_rights",
"acknowledge_privacy",
"idempotency_key"
],
"properties": {
"files": {
"type": "array",
"items": {
"type": "object",
"required": [
"field",
"filename",
"mime_type",
"base64"
],
"properties": {
"field": {
"type": "string"
},
"base64": {
"type": "string"
},
"filename": {
"type": "string"
},
"mime_type": {
"type": "string"
}
},
"additionalProperties": false
},
"maxItems": 1,
"description": "Optional upload parts (max 20 MiB total). Use audio for Voice, file for Edit/Upscale, image_upload/last_image_upload/voice_upload/music_upload for Video."
},
"uploads": {
"type": "array",
"items": {
"type": "object",
"title": "UploadInput",
"required": [
"field",
"asset_id"
],
"properties": {
"field": {
"type": "string",
"title": "Field"
},
"asset_id": {
"type": "string",
"title": "Asset Id",
"format": "uuid"
}
},
"additionalProperties": false
},
"maxItems": 1,
"minItems": 1,
"description": "Private uploaded file references from asset_upload or the web file library. No local paths, URLs or base64 in plans. Set the corresponding source/image/voice/music mode to upload."
},
"source_url": {
"type": [
"string",
"null"
],
"default": null,
"maxLength": 2048,
"description": "For source_type=url only: a public HTTPS image URL, max 2048 characters; no credentials, IP literals, non-443 ports or private network destinations."
},
"source_type": {
"enum": [
"upload",
"url",
"generated"
],
"type": "string",
"description": "Choose an uploaded source or a previous generated image."
},
"idempotency_key": {
"type": "string",
"maxLength": 128,
"minLength": 8,
"description": "Unique per intended operation. Reuse on network retries."
},
"generated_service": {
"enum": [
"image",
"edit",
null
],
"type": [
"string",
"null"
],
"default": null,
"description": "For generated input: image or edit. Omission defaults to image."
},
"generated_variant": {
"enum": [
"base",
"nsfw",
"result",
null
],
"type": [
"string",
"null"
],
"default": null,
"description": "For generated input: image uses base/nsfw; edit uses result."
},
"acknowledge_rights": {
"type": "boolean",
"const": true,
"description": "Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent."
},
"acknowledge_privacy": {
"type": "boolean",
"const": true,
"description": "Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent."
},
"generated_request_id": {
"type": [
"string",
"null"
],
"format": "uuid",
"default": null,
"maxLength": 36,
"description": "Owned completed source request UUID; required with source_type=generated."
}
},
"description": "Upscale one image from an upload, a public HTTPS image URL, or one owned Image/Edit result. Select exactly one source type.",
"additionalProperties": false
},
"annotations": {
"title": "Upscale image",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Submit voice work for the authenticated account. Returns an asynchronous request ID. For a sequence of work, prefer batch_submit and hand off the dashboard URL without polling. Use library IDs from assets_list where supported. Rights/privacy acknowledgements must reflect the user’s authorization; never invent consent. Limits remain active through result email delivery. Creative baseline: unless the user declined, build at least one dopamine hook from dopamine_hooks_get into this prompt, script or sound.
| Parameter | Type | Description |
|---|---|---|
| mode * | string | base requires reference audio + exact transcript + voice rights. custom uses a named speaker; design uses instruction. |
| text * | string | Text to speak, 1–2000 characters after trim and line-ending normalization. |
| files | array | Optional upload parts (max 20 MiB total). Use audio for Voice, file for Edit/Upscale, image_upload/last_image_upload/voice_upload/music_upload for Video. |
| speaker | string | Required in custom mode. Leave empty/omit in base and design. Names are case-sensitive. |
| uploads | array | Private uploaded file references from asset_upload or the web file library. No local paths, URLs or base64 in plans. Set the corresponding source/image/voice/music mode to upload. |
| language | string | Use the full case-sensitive language name, e.g. Korean, not ko. |
| instruction | string | Voice/tone instructions up to 1000 characters. design requires at least 3 characters; custom optionally accepts instructions; base forbids them. |
| reference_text | string | Exact transcript of the reference audio, 1–4000 characters in base mode. Leave empty/omit in custom/design. |
| idempotency_key * | string | Unique per intended operation. Reuse on network retries. |
| acknowledge_safety * | boolean | Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent. |
| reference_voice_id | string | Base clone only: owned completed Voice result, or bind a preceding Voice output id. Mutually exclusive with uploads/inline audio. The server uses the source text as the exact reference transcript; omit reference_text or supply the same text. No download/re-upload is needed. |
| acknowledge_privacy * | boolean | Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent. |
| acknowledge_voice_rights | boolean | Must be true in base mode after user authorization to reproduce the voice; never invent the speaker consent. |
{
"name": "voice_create",
"title": "Create voice",
"description": "Submit voice work for the authenticated account. Returns an asynchronous request ID. For a sequence of work, prefer batch_submit and hand off the dashboard URL without polling. Use library IDs from assets_list where supported. Rights/privacy acknowledgements must reflect the user’s authorization; never invent consent. Limits remain active through result email delivery. Creative baseline: unless the user declined, build at least one dopamine hook from dopamine_hooks_get into this prompt, script or sound.",
"inputSchema": {
"type": "object",
"allOf": [
{
"if": {
"required": [
"mode"
],
"properties": {
"mode": {
"const": "base"
}
}
},
"then": {
"required": [
"acknowledge_voice_rights"
],
"properties": {
"speaker": {
"maxLength": 0
},
"instruction": {
"maxLength": 0
},
"acknowledge_voice_rights": {
"const": true
}
}
}
},
{
"if": {
"required": [
"mode"
],
"properties": {
"mode": {
"const": "custom"
}
}
},
"then": {
"required": [
"speaker"
],
"properties": {
"speaker": {
"minLength": 1
},
"reference_text": {
"maxLength": 0
}
}
}
},
{
"if": {
"required": [
"mode"
],
"properties": {
"mode": {
"const": "design"
}
}
},
"then": {
"required": [
"instruction"
],
"properties": {
"speaker": {
"maxLength": 0
},
"instruction": {
"minLength": 3
},
"reference_text": {
"maxLength": 0
}
}
}
}
],
"required": [
"mode",
"text",
"acknowledge_privacy",
"acknowledge_safety",
"idempotency_key"
],
"properties": {
"mode": {
"enum": [
"base",
"custom",
"design"
],
"type": "string",
"description": "base requires reference audio + exact transcript + voice rights. custom uses a named speaker; design uses instruction."
},
"text": {
"type": "string",
"maxLength": 2000,
"minLength": 1,
"description": "Text to speak, 1–2000 characters after trim and line-ending normalization."
},
"files": {
"type": "array",
"items": {
"type": "object",
"required": [
"field",
"filename",
"mime_type",
"base64"
],
"properties": {
"field": {
"type": "string"
},
"base64": {
"type": "string"
},
"filename": {
"type": "string"
},
"mime_type": {
"type": "string"
}
},
"additionalProperties": false
},
"maxItems": 1,
"description": "Optional upload parts (max 20 MiB total). Use audio for Voice, file for Edit/Upscale, image_upload/last_image_upload/voice_upload/music_upload for Video."
},
"speaker": {
"enum": [
"",
"Sohee",
"Vivian",
"Serena",
"Uncle_Fu",
"Dylan",
"Eric",
"Ryan",
"Aiden",
"Ono_Anna"
],
"type": "string",
"default": "",
"examples": [
"Sohee"
],
"maxLength": 80,
"description": "Required in custom mode. Leave empty/omit in base and design. Names are case-sensitive."
},
"uploads": {
"type": "array",
"items": {
"type": "object",
"title": "UploadInput",
"required": [
"field",
"asset_id"
],
"properties": {
"field": {
"type": "string",
"title": "Field"
},
"asset_id": {
"type": "string",
"title": "Asset Id",
"format": "uuid"
}
},
"additionalProperties": false
},
"maxItems": 1,
"minItems": 1,
"description": "Private uploaded file references from asset_upload or the web file library. No local paths, URLs or base64 in plans. Set the corresponding source/image/voice/music mode to upload."
},
"language": {
"enum": [
"Korean",
"English",
"Chinese",
"Japanese",
"German",
"French",
"Russian",
"Portuguese",
"Spanish",
"Italian"
],
"type": "string",
"default": "Korean",
"examples": [
"Korean"
],
"maxLength": 40,
"description": "Use the full case-sensitive language name, e.g. Korean, not ko."
},
"instruction": {
"type": "string",
"default": "",
"maxLength": 1000,
"description": "Voice/tone instructions up to 1000 characters. design requires at least 3 characters; custom optionally accepts instructions; base forbids them."
},
"reference_text": {
"type": "string",
"default": "",
"maxLength": 4000,
"description": "Exact transcript of the reference audio, 1–4000 characters in base mode. Leave empty/omit in custom/design."
},
"idempotency_key": {
"type": "string",
"maxLength": 128,
"minLength": 8,
"description": "Unique per intended operation. Reuse on network retries."
},
"acknowledge_safety": {
"type": "boolean",
"const": true,
"description": "Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent."
},
"reference_voice_id": {
"type": "string",
"format": "uuid",
"description": "Base clone only: owned completed Voice result, or bind a preceding Voice output id. Mutually exclusive with uploads/inline audio. The server uses the source text as the exact reference transcript; omit reference_text or supply the same text. No download/re-upload is needed."
},
"acknowledge_privacy": {
"type": "boolean",
"const": true,
"description": "Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent."
},
"acknowledge_voice_rights": {
"type": "boolean",
"default": false,
"description": "Must be true in base mode after user authorization to reproduce the voice; never invent the speaker consent."
}
},
"description": "Generate speech. Language and speaker names are case-sensitive. base requires authorized 2–30-second reference audio and its exact transcript; custom selects a named speaker; design describes a voice. Base may instead use reference_voice_id with automatic source transcript reuse.",
"additionalProperties": false
},
"annotations": {
"title": "Create voice",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Submit music work for the authenticated account. Returns an asynchronous request ID. For a sequence of work, prefer batch_submit and hand off the dashboard URL without polling. Use library IDs from assets_list where supported. Rights/privacy acknowledgements must reflect the user’s authorization; never invent consent. Limits remain active through result email delivery. Creative baseline: unless the user declined, build at least one dopamine hook from dopamine_hooks_get into this prompt, script or sound.
| Parameter | Type | Description |
|---|---|---|
| bpm | integer | Integer tempo from30 through300 BPM; default120. |
| lyrics | string | Lyrics up to 4096 characters. Required and non-empty for instrumental=false; omit or send an empty string for instrumental=true. |
| language | string | Lyrics language code; unlike Voice, Music uses short codes such as ko. |
| key_scale | string | Case-sensitive note plus Major or Minor. Sharps are supported; flat spellings are not accepted. |
| description * | string | Music description, 3–2000 characters after trim and newline normalization. Do not include unsupported control characters. |
| instrumental | boolean | Set true for instrumental BGM and omit lyrics. Defaults false, which requires lyrics. |
| time_signature | string | API meter code: "2", "3", "4", or "6" only. For 4/4 time submit "4". Never send "4/4", "3/4" or "6/8" in this field; musical wording may appear in description. |
| idempotency_key * | string | Unique per intended operation. Reuse on network retries. |
| duration_seconds | integer | Exactly one of 10,15,30,60,120 seconds; default30. A 15-second ad can request15. |
| acknowledge_rights * | boolean | Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent. |
| acknowledge_safety * | boolean | Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent. |
| acknowledge_privacy * | boolean | Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent. |
{
"name": "music_create",
"title": "Create music",
"description": "Submit music work for the authenticated account. Returns an asynchronous request ID. For a sequence of work, prefer batch_submit and hand off the dashboard URL without polling. Use library IDs from assets_list where supported. Rights/privacy acknowledgements must reflect the user’s authorization; never invent consent. Limits remain active through result email delivery. Creative baseline: unless the user declined, build at least one dopamine hook from dopamine_hooks_get into this prompt, script or sound.",
"inputSchema": {
"type": "object",
"allOf": [
{
"if": {
"required": [
"instrumental"
],
"properties": {
"instrumental": {
"const": true
}
}
},
"else": {
"required": [
"lyrics"
],
"properties": {
"lyrics": {
"minLength": 1
}
}
},
"then": {
"properties": {
"lyrics": {
"maxLength": 0
}
}
}
}
],
"required": [
"description",
"acknowledge_rights",
"acknowledge_privacy",
"acknowledge_safety",
"idempotency_key"
],
"properties": {
"bpm": {
"type": "integer",
"default": 120,
"maximum": 300,
"minimum": 30,
"examples": [
120
],
"description": "Integer tempo from30 through300 BPM; default120."
},
"lyrics": {
"type": "string",
"default": "",
"maxLength": 4096,
"description": "Lyrics up to 4096 characters. Required and non-empty for instrumental=false; omit or send an empty string for instrumental=true."
},
"language": {
"enum": [
"ko",
"en",
"zh",
"ja",
"es",
"fr",
"de"
],
"type": "string",
"default": "ko",
"maxLength": 2,
"description": "Lyrics language code; unlike Voice, Music uses short codes such as ko."
},
"key_scale": {
"enum": [
"C Major",
"C Minor",
"C# Major",
"C# Minor",
"D Major",
"D Minor",
"D# Major",
"D# Minor",
"E Major",
"E Minor",
"F Major",
"F Minor",
"F# Major",
"F# Minor",
"G Major",
"G Minor",
"G# Major",
"G# Minor",
"A Major",
"A Minor",
"A# Major",
"A# Minor",
"B Major",
"B Minor"
],
"type": "string",
"default": "C Major",
"examples": [
"C Major",
"F# Minor"
],
"maxLength": 40,
"minLength": 3,
"description": "Case-sensitive note plus Major or Minor. Sharps are supported; flat spellings are not accepted."
},
"description": {
"type": "string",
"examples": [
"Original 15-second playful electro-pop instrumental, punchy first beat and a clean ending."
],
"maxLength": 2000,
"minLength": 3,
"description": "Music description, 3–2000 characters after trim and newline normalization. Do not include unsupported control characters."
},
"instrumental": {
"type": "boolean",
"default": false,
"description": "Set true for instrumental BGM and omit lyrics. Defaults false, which requires lyrics."
},
"time_signature": {
"enum": [
"2",
"3",
"4",
"6"
],
"type": "string",
"default": "4",
"examples": [
"4"
],
"maxLength": 2,
"description": "API meter code: \"2\", \"3\", \"4\", or \"6\" only. For 4/4 time submit \"4\". Never send \"4/4\", \"3/4\" or \"6/8\" in this field; musical wording may appear in description."
},
"idempotency_key": {
"type": "string",
"maxLength": 128,
"minLength": 8,
"description": "Unique per intended operation. Reuse on network retries."
},
"duration_seconds": {
"enum": [
10,
15,
30,
60,
120
],
"type": "integer",
"default": 30,
"maximum": 120,
"minimum": 10,
"description": "Exactly one of 10,15,30,60,120 seconds; default30. A 15-second ad can request15."
},
"acknowledge_rights": {
"type": "boolean",
"const": true,
"description": "Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent."
},
"acknowledge_safety": {
"type": "boolean",
"const": true,
"description": "Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent."
},
"acknowledge_privacy": {
"type": "boolean",
"const": true,
"description": "Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent."
}
},
"description": "Generate music. time_signature uses one of \"2\", \"3\", \"4\", \"6\" (for 4/4 submit \"4\", not \"4/4\"). Use instrumental=true with no lyrics, or instrumental=false with non-empty lyrics. Validate the final plan before submission.",
"additionalProperties": false
},
"annotations": {
"title": "Create music",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Submit sfx work for the authenticated account. Returns an asynchronous request ID. For a sequence of work, prefer batch_submit and hand off the dashboard URL without polling. Use library IDs from assets_list where supported. Rights/privacy acknowledgements must reflect the user’s authorization; never invent consent. Limits remain active through result email delivery. Creative baseline: unless the user declined, build at least one dopamine hook from dopamine_hooks_get into this prompt, script or sound. HiruAI has recorded provider acceptance of Stable Audio and Gemma terms for hosted SFX. Omit legacy acknowledge_model_terms; do not ask for a separate end-user model-terms confirmation. Rights, privacy and safety acknowledgements and model use restrictions remain in force.
| Parameter | Type | Description |
|---|---|---|
| seed | any | Optional random seed for the sound effect. |
| description * | string | Describe the sound effect to generate, including audible events and timing. |
| idempotency_key * | string | Unique per intended operation. Reuse on network retries. |
| duration_seconds | integer | Requested sound effect duration in seconds. |
| acknowledge_rights * | boolean | Confirm the user has the rights or permissions to use the source material. |
| acknowledge_safety * | boolean | Confirm the user accepts the safety restrictions for this request. |
| acknowledge_privacy * | boolean | Confirm the user has permission to process any personal data in the inputs. |
| acknowledge_model_terms | boolean | Optional legacy user acknowledgement. HiruAI has recorded provider acceptance of Stable Audio and Gemma terms for this hosted SFX service (2026-11-19-hiruai-sa3-gemma-provider-v1). Omit this field; do not ask the user to accept these model terms for each generation. Use restrictions still apply. |
{
"name": "sfx_create",
"title": "Generate sound effects",
"description": "Submit sfx work for the authenticated account. Returns an asynchronous request ID. For a sequence of work, prefer batch_submit and hand off the dashboard URL without polling. Use library IDs from assets_list where supported. Rights/privacy acknowledgements must reflect the user’s authorization; never invent consent. Limits remain active through result email delivery. Creative baseline: unless the user declined, build at least one dopamine hook from dopamine_hooks_get into this prompt, script or sound. HiruAI has recorded provider acceptance of Stable Audio and Gemma terms for hosted SFX. Omit legacy acknowledge_model_terms; do not ask for a separate end-user model-terms confirmation. Rights, privacy and safety acknowledgements and model use restrictions remain in force.",
"inputSchema": {
"type": "object",
"title": "SfxGenerationRequest",
"required": [
"description",
"acknowledge_rights",
"acknowledge_privacy",
"acknowledge_safety",
"idempotency_key"
],
"properties": {
"seed": {
"anyOf": [
{
"type": "integer",
"maximum": 4294967295,
"minimum": 0
},
{
"type": "null"
}
],
"title": "Seed",
"default": null,
"description": "Optional random seed for the sound effect."
},
"description": {
"type": "string",
"title": "Description",
"maxLength": 2000,
"minLength": 3,
"description": "Describe the sound effect to generate, including audible events and timing."
},
"idempotency_key": {
"type": "string",
"maxLength": 128,
"minLength": 8,
"description": "Unique per intended operation. Reuse on network retries."
},
"duration_seconds": {
"enum": [
1,
2,
3,
5,
10,
15,
20,
30
],
"type": "integer",
"title": "Duration Seconds",
"default": 5,
"maximum": 30,
"minimum": 1,
"description": "Requested sound effect duration in seconds."
},
"acknowledge_rights": {
"type": "boolean",
"title": "Acknowledge Rights",
"description": "Confirm the user has the rights or permissions to use the source material."
},
"acknowledge_safety": {
"type": "boolean",
"title": "Acknowledge Safety",
"description": "Confirm the user accepts the safety restrictions for this request."
},
"acknowledge_privacy": {
"type": "boolean",
"title": "Acknowledge Privacy",
"description": "Confirm the user has permission to process any personal data in the inputs."
},
"acknowledge_model_terms": {
"type": "boolean",
"title": "Acknowledge Model Terms",
"default": false,
"deprecated": true,
"description": "Optional legacy user acknowledgement. HiruAI has recorded provider acceptance of Stable Audio and Gemma terms for this hosted SFX service (2026-11-19-hiruai-sa3-gemma-provider-v1). Omit this field; do not ask the user to accept these model terms for each generation. Use restrictions still apply."
}
},
"description": "Generate a sound effect with Stable Audio 3 Medium. English sound description recommended; fixed 8 steps / CFG 1, 44.1kHz stereo WAV. HiruAI's recorded provider acceptance covers the hosted model terms; do not request a separate end-user model-terms acknowledgement. Rights, privacy and safety acknowledgements remain required, and model use restrictions still apply. No speech cloning or music controls.",
"additionalProperties": false
},
"annotations": {
"title": "Generate sound effects",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Submit video work for the authenticated account. Returns an asynchronous request ID. For a sequence of work, prefer batch_submit and hand off the dashboard URL without polling. Use library IDs from assets_list where supported. Rights/privacy acknowledgements must reflect the user’s authorization; never invent consent. Limits remain active through result email delivery. Creative baseline: unless the user declined, build at least one dopamine hook from dopamine_hooks_get into this prompt, script or sound.
| Parameter | Type | Description |
|---|---|---|
| files | array | Optional upload parts (max 20 MiB total). Use audio for Voice, file for Edit/Upscale, image_upload/last_image_upload/voice_upload/music_upload for Video. |
| width * | integer | Allowed pairs: 1920×1088,1088×1920,1024×1024. |
| height * | integer | Allowed pairs: 1920×1088,1088×1920,1024×1024. |
| prompt * | string | Motion/scene prompt, 3–2000 characters after trim and newline normalization; unsupported control characters are rejected. |
| uploads | array | Private uploaded file references from asset_upload or the web file library. No local paths, URLs or base64 in plans. Set the corresponding source/image/voice/music mode to upload. |
| voice_fit | string | Voice longer than the clip's audio window ((num_frames-1)/24 s, 2-5 s): trim keeps only the first window (published default); reject refuses before any request/quota so a full-preservation intent never loses speech. For narration longer than one clip, keep the full Voice in Movie (sync_plan.mode=con… |
| image_mode | string | Choose how the first image is supplied. |
| music_mode | string | Choose how optional background music is supplied. |
| num_frames * | integer | 24fps frame count; about 2/3/4/5 seconds. |
| voice_mode | string | Choose how an optional narration track is supplied. |
| image_service | ["string","null"] | Required for library mode: image, edit or upscale. |
| image_variant | ["string","null"] | Required for library mode: image base/nsfw, edit result, upscale upscaled/4k. |
| reference_mode | string | keyframes=RGB endpoint guidance. depth/canny/pose=full-video IC-LoRA control; inspect video_reference_capabilities first, unavailable modes fail without submitting a job. |
| idempotency_key * | string | Unique per intended operation. Reuse on network retries. |
| last_image_mode | string | Choose how the final image is supplied. |
| image_request_id | ["string","null"] | Required for library mode: owned completed request UUID, or bind a prior output ID. |
| music_request_id | ["string","null"] | Required for library mode: owned completed music request UUID. |
| voice_request_id | ["string","null"] | Required for library mode: owned completed voice request UUID. |
| acknowledge_rights * | boolean | Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent. |
| acknowledge_safety * | boolean | Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent. |
| last_image_service | ["string","null"] | Required for library mode: image, edit or upscale. |
| last_image_variant | ["string","null"] | Required for library mode: image base/nsfw, edit result, upscale upscaled/4k. |
| reference_movie_id | ["string","null"] | Completed owned Blender reference Movie ID. Bind the preceding movie step output id here. |
| reference_strength | number | Set the strength of the full-video conditioning reference. |
| acknowledge_privacy * | boolean | Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent. |
| last_image_request_id | ["string","null"] | Required for library mode: owned completed request UUID, or bind a prior output ID. |
{
"name": "video_create",
"title": "Create video clip",
"description": "Submit video work for the authenticated account. Returns an asynchronous request ID. For a sequence of work, prefer batch_submit and hand off the dashboard URL without polling. Use library IDs from assets_list where supported. Rights/privacy acknowledgements must reflect the user’s authorization; never invent consent. Limits remain active through result email delivery. Creative baseline: unless the user declined, build at least one dopamine hook from dopamine_hooks_get into this prompt, script or sound.",
"inputSchema": {
"type": "object",
"allOf": [
{
"oneOf": [
{
"properties": {
"width": {
"const": 1920
},
"height": {
"const": 1088
}
}
},
{
"properties": {
"width": {
"const": 1088
},
"height": {
"const": 1920
}
}
},
{
"properties": {
"width": {
"const": 1024
},
"height": {
"const": 1024
}
}
}
]
},
{
"if": {
"required": [
"image_mode"
],
"properties": {
"image_mode": {
"const": "library"
}
}
},
"else": {
"properties": {
"image_service": {
"type": "null"
},
"image_variant": {
"type": "null"
},
"image_request_id": {
"type": "null"
}
}
},
"then": {
"required": [
"image_service",
"image_request_id",
"image_variant"
],
"properties": {
"image_service": {
"type": "string"
},
"image_variant": {
"type": "string"
},
"image_request_id": {
"type": "string"
}
}
}
},
{
"if": {
"required": [
"image_service"
],
"properties": {
"image_service": {
"const": "image"
}
}
},
"then": {
"properties": {
"image_variant": {
"enum": [
"base",
"nsfw"
]
}
}
}
},
{
"if": {
"required": [
"image_service"
],
"properties": {
"image_service": {
"const": "edit"
}
}
},
"then": {
"properties": {
"image_variant": {
"enum": [
"result"
]
}
}
}
},
{
"if": {
"required": [
"image_service"
],
"properties": {
"image_service": {
"const": "upscale"
}
}
},
"then": {
"properties": {
"image_variant": {
"enum": [
"upscaled",
"4k"
]
}
}
}
},
{
"if": {
"required": [
"last_image_mode"
],
"properties": {
"last_image_mode": {
"const": "library"
}
}
},
"else": {
"properties": {
"last_image_service": {
"type": "null"
},
"last_image_variant": {
"type": "null"
},
"last_image_request_id": {
"type": "null"
}
}
},
"then": {
"required": [
"last_image_service",
"last_image_request_id",
"last_image_variant"
],
"properties": {
"last_image_service": {
"type": "string"
},
"last_image_variant": {
"type": "string"
},
"last_image_request_id": {
"type": "string"
}
}
}
},
{
"if": {
"required": [
"last_image_service"
],
"properties": {
"last_image_service": {
"const": "image"
}
}
},
"then": {
"properties": {
"last_image_variant": {
"enum": [
"base",
"nsfw"
]
}
}
}
},
{
"if": {
"required": [
"last_image_service"
],
"properties": {
"last_image_service": {
"const": "edit"
}
}
},
"then": {
"properties": {
"last_image_variant": {
"enum": [
"result"
]
}
}
}
},
{
"if": {
"required": [
"last_image_service"
],
"properties": {
"last_image_service": {
"const": "upscale"
}
}
},
"then": {
"properties": {
"last_image_variant": {
"enum": [
"upscaled",
"4k"
]
}
}
}
},
{
"if": {
"required": [
"voice_mode"
],
"properties": {
"voice_mode": {
"const": "library"
}
}
},
"then": {
"required": [
"voice_request_id"
],
"properties": {
"voice_request_id": {
"type": "string"
}
}
}
},
{
"if": {
"required": [
"voice_mode"
],
"properties": {
"voice_mode": {
"const": "none"
}
}
},
"then": {
"properties": {
"voice_request_id": {
"type": "null"
}
}
}
},
{
"if": {
"required": [
"music_mode"
],
"properties": {
"music_mode": {
"const": "library"
}
}
},
"then": {
"required": [
"music_request_id"
],
"properties": {
"music_request_id": {
"type": "string"
}
}
}
},
{
"if": {
"required": [
"music_mode"
],
"properties": {
"music_mode": {
"const": "none"
}
}
},
"then": {
"properties": {
"music_request_id": {
"type": "null"
}
}
}
},
{
"if": {
"required": [
"last_image_mode"
],
"properties": {
"last_image_mode": {
"enum": [
"upload",
"library"
]
}
}
},
"then": {
"required": [
"image_mode"
],
"properties": {
"image_mode": {
"enum": [
"upload",
"library"
]
}
}
}
}
],
"required": [
"prompt",
"width",
"height",
"num_frames",
"acknowledge_rights",
"acknowledge_privacy",
"acknowledge_safety",
"idempotency_key"
],
"properties": {
"files": {
"type": "array",
"items": {
"type": "object",
"required": [
"field",
"filename",
"mime_type",
"base64"
],
"properties": {
"field": {
"type": "string"
},
"base64": {
"type": "string"
},
"filename": {
"type": "string"
},
"mime_type": {
"type": "string"
}
},
"additionalProperties": false
},
"maxItems": 4,
"description": "Optional upload parts (max 20 MiB total). Use audio for Voice, file for Edit/Upscale, image_upload/last_image_upload/voice_upload/music_upload for Video."
},
"width": {
"enum": [
1024,
1088,
1920
],
"type": "integer",
"description": "Allowed pairs: 1920×1088,1088×1920,1024×1024."
},
"height": {
"enum": [
1024,
1088,
1920
],
"type": "integer",
"description": "Allowed pairs: 1920×1088,1088×1920,1024×1024."
},
"prompt": {
"type": "string",
"maxLength": 2000,
"minLength": 3,
"description": "Motion/scene prompt, 3–2000 characters after trim and newline normalization; unsupported control characters are rejected."
},
"uploads": {
"type": "array",
"items": {
"type": "object",
"title": "UploadInput",
"required": [
"field",
"asset_id"
],
"properties": {
"field": {
"type": "string",
"title": "Field"
},
"asset_id": {
"type": "string",
"title": "Asset Id",
"format": "uuid"
}
},
"additionalProperties": false
},
"maxItems": 4,
"minItems": 1,
"description": "Private uploaded file references from asset_upload or the web file library. No local paths, URLs or base64 in plans. Set the corresponding source/image/voice/music mode to upload."
},
"voice_fit": {
"enum": [
"trim",
"reject"
],
"type": "string",
"default": "trim",
"description": "Voice longer than the clip's audio window ((num_frames-1)/24 s, 2-5 s): trim keeps only the first window (published default); reject refuses before any request/quota so a full-preservation intent never loses speech. For narration longer than one clip, keep the full Voice in Movie (sync_plan.mode=continuous/scene_sources) and set the clip volume=0."
},
"image_mode": {
"enum": [
"none",
"upload",
"library"
],
"type": "string",
"default": "none",
"description": "Choose how the first image is supplied."
},
"music_mode": {
"enum": [
"none",
"upload",
"library"
],
"type": "string",
"default": "none",
"description": "Choose how optional background music is supplied."
},
"num_frames": {
"enum": [
49,
73,
97,
121
],
"type": "integer",
"description": "24fps frame count; about 2/3/4/5 seconds."
},
"voice_mode": {
"enum": [
"none",
"upload",
"library"
],
"type": "string",
"default": "none",
"description": "Choose how an optional narration track is supplied."
},
"image_service": {
"enum": [
"image",
"edit",
"upscale",
null,
"movie"
],
"type": [
"string",
"null"
],
"default": null,
"description": "Required for library mode: image, edit or upscale."
},
"image_variant": {
"enum": [
"base",
"nsfw",
"result",
"upscaled",
"4k",
null,
"first",
"last"
],
"type": [
"string",
"null"
],
"default": null,
"description": "Required for library mode: image base/nsfw, edit result, upscale upscaled/4k."
},
"reference_mode": {
"enum": [
"none",
"keyframes",
"depth",
"canny",
"pose"
],
"type": "string",
"default": "none",
"description": "keyframes=RGB endpoint guidance. depth/canny/pose=full-video IC-LoRA control; inspect video_reference_capabilities first, unavailable modes fail without submitting a job."
},
"idempotency_key": {
"type": "string",
"maxLength": 128,
"minLength": 8,
"description": "Unique per intended operation. Reuse on network retries."
},
"last_image_mode": {
"enum": [
"none",
"upload",
"library"
],
"type": "string",
"default": "none",
"description": "Choose how the final image is supplied."
},
"image_request_id": {
"type": [
"string",
"null"
],
"format": "uuid",
"default": null,
"description": "Required for library mode: owned completed request UUID, or bind a prior output ID."
},
"music_request_id": {
"type": [
"string",
"null"
],
"format": "uuid",
"default": null,
"description": "Required for library mode: owned completed music request UUID."
},
"voice_request_id": {
"type": [
"string",
"null"
],
"format": "uuid",
"default": null,
"description": "Required for library mode: owned completed voice request UUID."
},
"acknowledge_rights": {
"type": "boolean",
"const": true,
"description": "Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent."
},
"acknowledge_safety": {
"type": "boolean",
"const": true,
"description": "Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent."
},
"last_image_service": {
"enum": [
"image",
"edit",
"upscale",
null,
"movie"
],
"type": [
"string",
"null"
],
"default": null,
"description": "Required for library mode: image, edit or upscale."
},
"last_image_variant": {
"enum": [
"base",
"nsfw",
"result",
"upscaled",
"4k",
null,
"first",
"last"
],
"type": [
"string",
"null"
],
"default": null,
"description": "Required for library mode: image base/nsfw, edit result, upscale upscaled/4k."
},
"reference_movie_id": {
"type": [
"string",
"null"
],
"format": "uuid",
"description": "Completed owned Blender reference Movie ID. Bind the preceding movie step output id here."
},
"reference_strength": {
"type": "number",
"default": 1,
"maximum": 1,
"minimum": 0,
"description": "Set the strength of the full-video conditioning reference."
},
"acknowledge_privacy": {
"type": "boolean",
"const": true,
"description": "Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent."
},
"last_image_request_id": {
"type": [
"string",
"null"
],
"format": "uuid",
"default": null,
"description": "Required for library mode: owned completed request UUID, or bind a prior output ID."
}
},
"description": "Generate 24fps video with one of three exact size pairs and 49/73/97/121 frames. Source modes use none, upload or owned library; never mix modes. A final reference image also requires a first reference image.",
"additionalProperties": false
},
"annotations": {
"title": "Create video clip",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Submit movie work for the authenticated account. Returns an asynchronous request ID. For a sequence of work, prefer batch_submit and hand off the dashboard URL without polling. Use library IDs from assets_list where supported. Rights/privacy acknowledgements must reflect the user’s authorization; never invent consent. Limits remain active through result email delivery. Creative baseline: unless the user declined, build at least one dopamine hook from dopamine_hooks_get into this prompt, script or sound.
| Parameter | Type | Description |
|---|---|---|
| clips | array | Ordered visual clips with their durations, sources and optional trims. |
| music | any | Optional background music track with placement on the movie timeline. |
| title | string | Display title for the plan or movie. |
| voice | any | Optional narration track with placement on the movie timeline. |
| aspect | string | Aspect ratio category for the movie output. |
| version | integer | Movie timeline schema version. |
| duration | any | Target movie duration in seconds. |
| graphics | array | Timed on-screen elements such as titles and captions. |
| sync_plan | any | Submit before Voice finishes. Bind a future voice asset; the server reads its output timing and resolves sentence/scene placement automatically after completion. |
| audio_layers | array | Optional audio tracks for voice, music and sound effects, with timing and gain. |
| blender_program | any | Optional Blender scene or timeline edit with its assets and output settings. |
| idempotency_key * | string | Unique per intended operation. Reuse on network retries. |
| acknowledge_rights | boolean | Confirm the user has the rights or permissions to use the source material. |
| acknowledge_privacy | boolean | Confirm the user has permission to process any personal data in the inputs. |
{
"name": "movie_create",
"title": "Compose movie",
"description": "Submit movie work for the authenticated account. Returns an asynchronous request ID. For a sequence of work, prefer batch_submit and hand off the dashboard URL without polling. Use library IDs from assets_list where supported. Rights/privacy acknowledgements must reflect the user’s authorization; never invent consent. Limits remain active through result email delivery. Creative baseline: unless the user declined, build at least one dopamine hook from dopamine_hooks_get into this prompt, script or sound.",
"inputSchema": {
"type": "object",
"$defs": {
"Cue": {
"type": "object",
"title": "Cue",
"required": [
"start",
"end",
"text"
],
"properties": {
"end": {
"type": "number",
"title": "End",
"maximum": 180,
"description": "Seconds in the original audio, exclusive.",
"exclusiveMinimum": 0
},
"text": {
"type": "string",
"title": "Text",
"maxLength": 160,
"minLength": 1
},
"start": {
"type": "number",
"title": "Start",
"minimum": 0,
"description": "Seconds in the original audio, inclusive.",
"exclusiveMaximum": 180
}
},
"additionalProperties": false
},
"Clip": {
"type": "object",
"title": "Clip",
"required": [
"asset",
"duration"
],
"properties": {
"trim": {
"type": "number",
"title": "Trim",
"default": 0,
"maximum": 180,
"minimum": 0
},
"asset": {
"$ref": "#/$defs/AssetRef"
},
"volume": {
"type": "number",
"title": "Volume",
"default": 1,
"maximum": 2,
"minimum": 0
},
"duration": {
"type": "number",
"title": "Duration",
"maximum": 120,
"minimum": 0.25
},
"hold_last_frame": {
"type": "boolean",
"title": "Hold Last Frame",
"default": false,
"description": "Allow extending a video by holding its final frame and padding audio with silence."
}
},
"additionalProperties": false
},
"AssetRef": {
"type": "object",
"title": "AssetRef",
"required": [
"service",
"id"
],
"properties": {
"id": {
"type": "string",
"title": "Id",
"format": "uuid"
},
"service": {
"enum": [
"upload",
"video",
"movie",
"image",
"edit",
"upscale",
"voice",
"music",
"sfx"
],
"type": "string",
"title": "Service"
},
"variant": {
"enum": [
"base",
"nsfw",
"result",
"upscaled",
"4k",
"first",
"last"
],
"type": "string",
"title": "Variant",
"default": "result"
}
},
"additionalProperties": false
},
"GainPoint": {
"type": "object",
"title": "GainPoint",
"required": [
"time",
"volume"
],
"properties": {
"time": {
"type": "number",
"title": "Time",
"maximum": 120,
"minimum": 0,
"description": "Seconds relative to the layer start."
},
"volume": {
"type": "number",
"title": "Volume",
"maximum": 2,
"minimum": 0
}
},
"additionalProperties": false
},
"AudioLayer": {
"type": "object",
"title": "AudioLayer",
"required": [
"asset",
"id",
"source_duration",
"duration"
],
"properties": {
"id": {
"type": "string",
"title": "Id",
"pattern": "^[A-Za-z0-9_-]{1,64}$"
},
"cues": {
"type": "array",
"items": {
"$ref": "#/$defs/Cue"
},
"title": "Cues",
"maxItems": 120,
"description": "Verified/imported source timestamps; never infer speech alignment from text length."
},
"duck": {
"type": "boolean",
"title": "Duck",
"default": false
},
"fade": {
"type": "number",
"title": "Fade",
"default": 1,
"maximum": 5,
"minimum": 0
},
"loop": {
"type": "boolean",
"title": "Loop",
"default": false
},
"name": {
"type": "string",
"title": "Name",
"default": "오디오 레이어",
"maxLength": 120
},
"role": {
"enum": [
"voice",
"music",
"sfx"
],
"type": "string",
"title": "Role",
"default": "voice"
},
"trim": {
"type": "number",
"title": "Trim",
"default": 0,
"maximum": 180,
"minimum": 0
},
"asset": {
"$ref": "#/$defs/AssetRef",
"description": "Audio-only source: Voice, Music, SFX result or an uploaded audio file. Video/Movie assets belong in clips; use clips.volume for their sound or privately upload extracted audio. Uploaded media kind is checked at dispatch."
},
"muted": {
"type": "boolean",
"title": "Muted",
"default": false
},
"offset": {
"type": "number",
"title": "Offset",
"default": 0,
"minimum": 0,
"exclusiveMaximum": 120
},
"volume": {
"type": "number",
"title": "Volume",
"default": 1,
"maximum": 2,
"minimum": 0
},
"captions": {
"type": "boolean",
"title": "Captions",
"default": false
},
"duration": {
"type": "number",
"title": "Duration",
"maximum": 120,
"minimum": 0.03333333333333333,
"description": "Length on the movie timeline; no speed change. Non-looping layers cannot exceed source_duration."
},
"envelope": {
"type": "array",
"items": {
"$ref": "#/$defs/GainPoint"
},
"title": "Envelope",
"maxItems": 32
},
"clip_index": {
"anyOf": [
{
"type": "integer",
"maximum": 23,
"minimum": 0
},
{
"type": "null"
}
],
"title": "Clip Index",
"default": null,
"description": "Optional scene anchor. offset is relative to this scene start; resolved after measured narration changes scene lengths. Omit for absolute movie seconds."
},
"caption_preset": {
"enum": [
"fade",
"slide_up",
"pop",
"typewriter"
],
"type": "string",
"title": "Caption Preset",
"default": "fade"
},
"preserve_source": {
"type": "boolean",
"title": "Preserve Source",
"default": false,
"description": "Protect a complete narration source: trim=0, no loop/fade/envelope/mute, full measured file duration. Timeline Blender programs must preserve this audio and resolved visual boundaries."
},
"source_duration": {
"type": "number",
"title": "Source Duration",
"maximum": 180,
"description": "Length of the selected source region, not the entire file. Loop repeats only this region.",
"exclusiveMinimum": 0
}
},
"additionalProperties": false
},
"AudioTrack": {
"type": "object",
"title": "AudioTrack",
"required": [
"asset"
],
"properties": {
"fade": {
"type": "number",
"title": "Fade",
"default": 1,
"maximum": 5,
"minimum": 0
},
"loop": {
"type": "boolean",
"title": "Loop",
"default": false
},
"trim": {
"type": "number",
"title": "Trim",
"default": 0,
"maximum": 180,
"minimum": 0
},
"asset": {
"$ref": "#/$defs/AssetRef",
"description": "Audio-only source: Voice, Music, SFX result or an uploaded audio file. Video/Movie assets belong in clips; use clips.volume for their sound or privately upload extracted audio. Uploaded media kind is checked at dispatch."
},
"offset": {
"type": "number",
"title": "Offset",
"default": 0,
"minimum": 0,
"exclusiveMaximum": 120
},
"volume": {
"type": "number",
"title": "Volume",
"default": 1,
"maximum": 2,
"minimum": 0
}
},
"additionalProperties": false
},
"ProgramAsset": {
"type": "object",
"title": "ProgramAsset",
"required": [
"name",
"asset"
],
"properties": {
"name": {
"type": "string",
"title": "Name",
"pattern": "^[A-Za-z][A-Za-z0-9_-]{0,39}$"
},
"asset": {
"$ref": "#/$defs/AssetRef"
}
},
"additionalProperties": false
},
"MotionGraphic": {
"type": "object",
"title": "MotionGraphic",
"required": [
"id",
"text",
"start",
"end"
],
"properties": {
"x": {
"type": "number",
"title": "X",
"default": 0.5,
"maximum": 0.9,
"minimum": 0.1,
"description": "Normalized horizontal center, left to right."
},
"y": {
"type": "number",
"title": "Y",
"default": 0.5,
"maximum": 0.9,
"minimum": 0.1,
"description": "Normalized vertical center, top to bottom."
},
"id": {
"type": "string",
"title": "Id",
"pattern": "^[A-Za-z0-9_-]{1,64}$"
},
"end": {
"type": "number",
"title": "End",
"maximum": 120,
"exclusiveMinimum": 0
},
"size": {
"type": "number",
"title": "Size",
"default": 0.065,
"maximum": 0.15,
"minimum": 0.02,
"description": "Font size as fraction of output height."
},
"text": {
"type": "string",
"title": "Text",
"maxLength": 160,
"minLength": 1
},
"color": {
"type": "string",
"title": "Color",
"default": "#ffffff",
"pattern": "^#[0-9a-fA-F]{6}$"
},
"start": {
"type": "number",
"title": "Start",
"minimum": 0,
"exclusiveMaximum": 120
},
"preset": {
"enum": [
"fade",
"slide_up",
"pop",
"typewriter"
],
"type": "string",
"title": "Preset",
"default": "slide_up"
},
"animation": {
"type": "number",
"title": "Animation",
"default": 0.3,
"maximum": 2,
"minimum": 0
},
"clip_index": {
"anyOf": [
{
"type": "integer",
"maximum": 23,
"minimum": 0
},
{
"type": "null"
}
],
"title": "Clip Index",
"default": null,
"description": "When supplied, start/end are relative to this scene; resolved after narration timing."
},
"follow_clip": {
"type": "boolean",
"title": "Follow Clip",
"default": false,
"description": "With clip_index, extend the title until the actual end of that scene."
}
},
"additionalProperties": false
},
"NarrationSync": {
"type": "object",
"title": "NarrationSync",
"required": [
"segments"
],
"properties": {
"mode": {
"enum": [
"segments",
"continuous",
"scene_sources"
],
"type": "string",
"title": "Mode",
"default": "segments",
"description": "Audio policy. continuous: play ONE complete source once, preserve original pauses, cut visuals at measured sentence starts; requires all sentence IDs in order. scene_sources: each segment plays ITS OWN complete source file (scene Voice clone) once from sample 0 to the measured file end, never trimmed to ASR cue ends; requires all of that source's sentence IDs. segments: legacy SELECTED-REGION edit that trims each scene to ASR sentence spans with 20ms fades; it does not promise full-source coverage. Omitted mode means segments for backward compatibility; new plans should always state the mode."
},
"asset": {
"anyOf": [
{
"$ref": "#/$defs/AssetRef"
},
{
"type": "null"
}
],
"default": null,
"description": "Default narration for segments without their own asset. May be omitted when every segment has an asset."
},
"segments": {
"type": "array",
"items": {
"$ref": "#/$defs/NarrationSegment"
},
"title": "Segments",
"maxItems": 24,
"minItems": 1
},
"end_guard": {
"type": "number",
"title": "End Guard",
"default": 1,
"maximum": 3,
"minimum": 1,
"description": "continuous and scene_sources: mandatory silence after the actual complete file end of the final narrated scene; never shrunk. ASR sentence end is NOT an audio trim boundary."
},
"visual_fit": {
"enum": [
"hold",
"strict"
],
"type": "string",
"title": "Visual Fit",
"default": "hold"
},
"start_padding": {
"type": "number",
"title": "Start Padding",
"default": 0.15,
"maximum": 3,
"minimum": 0,
"description": "continuous only: silence before the whole source, after any unbound intro clips."
},
"padding_policy": {
"enum": [
"preserve",
"shrink"
],
"type": "string",
"title": "Padding Policy",
"default": "preserve",
"description": "preserve keeps requested lead/tail. With fixed duration, shrink may proportionally reduce only lead/tail to fit; speech is never cut or sped up. For exact 15s shorts use fixed + shrink. If speech alone exceeds the budget, shorten the script or select fewer sentences."
},
"duration_policy": {
"enum": [
"fixed",
"expand"
],
"type": "string",
"title": "Duration Policy",
"default": "expand",
"description": "expand chooses the next allowed duration that fits actual speech, up to 120 seconds. fixed stops if speech exceeds the selected duration."
}
},
"additionalProperties": false
},
"BlenderProgram": {
"type": "object",
"title": "BlenderProgram",
"required": [
"summary",
"python"
],
"properties": {
"mode": {
"enum": [
"scene",
"timeline"
],
"type": "string",
"title": "Mode",
"default": "scene",
"description": "scene builds a new 3D/VSE/compositor scene; timeline edits the prepared Movie VSE project including its timed audio and text."
},
"assets": {
"type": "array",
"items": {
"$ref": "#/$defs/ProgramAsset"
},
"title": "Assets",
"maxItems": 24
},
"python": {
"type": "string",
"title": "Python",
"maxLength": 100000,
"minLength": 1,
"description": "Blender 5.2.1 bpy Python. Runs in an isolated CPU container, never the web/worker process. Build or edit bpy.context.scene; server renders it. HIRU contains resolved timeline, assets and output dimensions. No network, credentials, external packages or GPU."
},
"purpose": {
"enum": [
"movie",
"video_reference"
],
"type": "string",
"title": "Purpose",
"default": "movie"
},
"summary": {
"type": "string",
"title": "Summary",
"maxLength": 2000,
"minLength": 10,
"description": "Human-readable creative intent, including camera, animation, compositing and audio changes. Shown in Movie history."
},
"reference_kind": {
"enum": [
"rgb",
"depth",
"canny",
"pose"
],
"type": "string",
"title": "Reference Kind",
"default": "rgb",
"description": "Representation rendered by this script. Non-RGB control maps require a matching verified LTX IC-LoRA path."
},
"timeline_layout": {
"enum": [
"joined",
"clips"
],
"type": "string",
"title": "Timeline Layout",
"default": "joined",
"description": "joined preserves the legacy single MOVIE strip. clips exposes each original photo as IMAGE and each prepared video as MOVIE, named clip-0, clip-1, etc. Use clips for per-photo zoom/pan; audio and captions stay separate."
},
"required_transform_clips": {
"type": "array",
"items": {
"type": "integer"
},
"title": "Required Transform Clips",
"maxItems": 24,
"description": "With timeline_layout=clips, clip indices whose scale/offset/rotation must change during their own frame interval. A missing, muted, or static target fails before rendering; use for promised photo zoom/pan. This is transform validation, not proof of arbitrary visual effects."
}
},
"additionalProperties": false
},
"NarrationSegment": {
"type": "object",
"title": "NarrationSegment",
"required": [
"clip_index",
"sentence_ids"
],
"properties": {
"lead": {
"type": "number",
"title": "Lead",
"default": 0.15,
"maximum": 3,
"minimum": 0
},
"tail": {
"type": "number",
"title": "Tail",
"default": 0.25,
"maximum": 3,
"minimum": 0
},
"asset": {
"anyOf": [
{
"$ref": "#/$defs/AssetRef"
},
{
"type": "null"
}
],
"default": null,
"description": "Optional per-scene narration. Overrides sync_plan.asset; sentence IDs belong to this source. Bind each scene clone output asset here."
},
"captions": {
"type": "boolean",
"title": "Captions",
"default": true
},
"clip_index": {
"type": "integer",
"title": "Clip Index",
"maximum": 23,
"minimum": 0
},
"sentence_ids": {
"type": "array",
"items": {
"type": "string"
},
"title": "Sentence Ids",
"maxItems": 120,
"minItems": 1,
"description": "Stable source transcript IDs s0, s1, ...; one sentence per line is recommended."
},
"caption_preset": {
"enum": [
"fade",
"slide_up",
"pop",
"typewriter"
],
"type": "string",
"title": "Caption Preset",
"default": "fade"
}
},
"additionalProperties": false
}
},
"title": "Timeline",
"required": [
"idempotency_key"
],
"properties": {
"clips": {
"type": "array",
"items": {
"$ref": "#/$defs/Clip"
},
"title": "Clips",
"maxItems": 24,
"description": "Ordered visual clips with their durations, sources and optional trims."
},
"music": {
"anyOf": [
{
"$ref": "#/$defs/AudioTrack"
},
{
"type": "null"
}
],
"default": null,
"description": "Optional background music track with placement on the movie timeline."
},
"title": {
"type": "string",
"title": "Title",
"default": "나의 짧은 영화",
"maxLength": 120,
"minLength": 1,
"description": "Display title for the plan or movie."
},
"voice": {
"anyOf": [
{
"$ref": "#/$defs/AudioTrack"
},
{
"type": "null"
}
],
"default": null,
"description": "Optional narration track with placement on the movie timeline."
},
"aspect": {
"enum": [
"landscape",
"portrait",
"square"
],
"type": "string",
"title": "Aspect",
"default": "landscape",
"description": "Aspect ratio category for the movie output."
},
"version": {
"enum": [
1,
2
],
"type": "integer",
"title": "Version",
"default": 2,
"description": "Movie timeline schema version."
},
"duration": {
"anyOf": [
{
"enum": [
2,
3,
4,
5
],
"type": "integer"
},
{
"type": "integer",
"maximum": 120,
"minimum": 10
}
],
"title": "Duration",
"default": 15,
"description": "Target movie duration in seconds."
},
"graphics": {
"type": "array",
"items": {
"$ref": "#/$defs/MotionGraphic"
},
"title": "Graphics",
"maxItems": 64,
"description": "Timed on-screen elements such as titles and captions."
},
"sync_plan": {
"anyOf": [
{
"$ref": "#/$defs/NarrationSync"
},
{
"type": "null"
}
],
"default": null,
"description": "Submit before Voice finishes. Bind a future voice asset; the server reads its output timing and resolves sentence/scene placement automatically after completion."
},
"audio_layers": {
"type": "array",
"items": {
"$ref": "#/$defs/AudioLayer"
},
"title": "Audio Layers",
"maxItems": 32,
"description": "Optional audio tracks for voice, music and sound effects, with timing and gain."
},
"blender_program": {
"anyOf": [
{
"$ref": "#/$defs/BlenderProgram"
},
{
"type": "null"
}
],
"default": null,
"description": "Optional Blender scene or timeline edit with its assets and output settings."
},
"idempotency_key": {
"type": "string",
"maxLength": 128,
"minLength": 8,
"description": "Unique per intended operation. Reuse on network retries."
},
"acknowledge_rights": {
"type": "boolean",
"title": "Acknowledge Rights",
"default": false,
"description": "Confirm the user has the rights or permissions to use the source material."
},
"acknowledge_privacy": {
"type": "boolean",
"title": "Acknowledge Privacy",
"default": false,
"description": "Confirm the user has permission to process any personal data in the inputs."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Compose movie",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Submit media work for the authenticated account. Returns an asynchronous request ID. For a sequence of work, prefer batch_submit and hand off the dashboard URL without polling. Use library IDs from assets_list where supported. Rights/privacy acknowledgements must reflect the user’s authorization; never invent consent. Limits remain active through result email delivery.
| Parameter | Type | Description |
|---|---|---|
| title | ["string","null"] | Optional request title, up to200 characters. |
| source_url * | string | Single YouTube/Shorts/Live/TikTok video HTTPS URL,10–2048 characters. No playlists, credentials or non-443 ports. The backend validates allowed hosts and paths. |
| idempotency_key * | string | Unique per intended operation. Reuse on network retries. |
| acknowledge_rights * | boolean | Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent. |
| acknowledge_privacy * | boolean | Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent. |
{
"name": "media_create",
"title": "Analyze public video",
"description": "Submit media work for the authenticated account. Returns an asynchronous request ID. For a sequence of work, prefer batch_submit and hand off the dashboard URL without polling. Use library IDs from assets_list where supported. Rights/privacy acknowledgements must reflect the user’s authorization; never invent consent. Limits remain active through result email delivery.",
"inputSchema": {
"type": "object",
"required": [
"source_url",
"acknowledge_rights",
"acknowledge_privacy",
"idempotency_key"
],
"properties": {
"title": {
"type": [
"string",
"null"
],
"default": null,
"maxLength": 200,
"description": "Optional request title, up to200 characters."
},
"source_url": {
"type": "string",
"maxLength": 2048,
"minLength": 10,
"description": "Single YouTube/Shorts/Live/TikTok video HTTPS URL,10–2048 characters. No playlists, credentials or non-443 ports. The backend validates allowed hosts and paths."
},
"idempotency_key": {
"type": "string",
"maxLength": 128,
"minLength": 8,
"description": "Unique per intended operation. Reuse on network retries."
},
"acknowledge_rights": {
"type": "boolean",
"const": true,
"description": "Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent."
},
"acknowledge_privacy": {
"type": "boolean",
"const": true,
"description": "Must be true only when the user has authorized the stated rights, privacy or safety acknowledgement. Never invent consent."
}
},
"description": "Analyze one permitted public HTTPS YouTube/Shorts/Live/TikTok video URL. Playlists, credentials, IP hosts, custom ports and other websites are unsupported.",
"additionalProperties": false
},
"annotations": {
"title": "Analyze public video",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
}Read the same image-model catalog as image.hiruai.com: current execution availability, traits, dated Arena evidence, model-specific consent and ChatGPT/Claude native-image alternatives. Show the relevant comparison to the user before selecting a model. This read-only call creates no job and consumes no generation quota.
No parameters.
{
"name": "image_models_get",
"title": "Compare image models",
"description": "Read the same image-model catalog as image.hiruai.com: current execution availability, traits, dated Arena evidence, model-specific consent and ChatGPT/Claude native-image alternatives. Show the relevant comparison to the user before selecting a model. This read-only call creates no job and consumes no generation quota.",
"inputSchema": {
"type": "object",
"properties": {},
"additionalProperties": false
},
"annotations": {
"title": "Compare image models",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Upload a local image, audio or video as base64 to your private file library for up to 30 days. If the new file exceeds the library’s space or file-count limit, the oldest uploads that unfinished plans no longer need are deleted first. Returns an upload asset ID and evicted_count without a GPU job. Use this ID in uploads for Video/Edit/Upscale/Voice, or as a Movie asset. Reuse idempotency_key on retries. Never put base64 or a local path in batch arguments.
| Parameter | Type | Description |
|---|---|---|
| base64 * | string | Source file bytes encoded as base64. |
| filename * | string | Filename for the source upload. |
| mime_type * | string | Media type of the uploaded file. |
| idempotency_key * | string | Unique key for the intended operation. Reuse it on network retries. |
| acknowledge_rights * | boolean | Confirm the user has the rights or permissions to use the source material. |
| acknowledge_privacy * | boolean | Confirm the user has permission to process any personal data in the inputs. |
{
"name": "asset_upload",
"title": "Upload file to private library",
"description": "Upload a local image, audio or video as base64 to your private file library for up to 30 days. If the new file exceeds the library’s space or file-count limit, the oldest uploads that unfinished plans no longer need are deleted first. Returns an upload asset ID and evicted_count without a GPU job. Use this ID in uploads for Video/Edit/Upscale/Voice, or as a Movie asset. Reuse idempotency_key on retries. Never put base64 or a local path in batch arguments.",
"inputSchema": {
"type": "object",
"required": [
"filename",
"mime_type",
"base64",
"idempotency_key",
"acknowledge_rights",
"acknowledge_privacy"
],
"properties": {
"base64": {
"type": "string",
"maxLength": 27962028,
"description": "Source file bytes encoded as base64."
},
"filename": {
"type": "string",
"maxLength": 200,
"minLength": 1,
"description": "Filename for the source upload."
},
"mime_type": {
"enum": [
"image/png",
"image/jpeg",
"image/webp",
"audio/wav",
"audio/mpeg",
"audio/mp4",
"video/mp4",
"video/webm"
],
"type": "string",
"description": "Media type of the uploaded file."
},
"idempotency_key": {
"type": "string",
"maxLength": 128,
"minLength": 8,
"description": "Unique key for the intended operation. Reuse it on network retries."
},
"acknowledge_rights": {
"type": "boolean",
"description": "Confirm the user has the rights or permissions to use the source material."
},
"acknowledge_privacy": {
"type": "boolean",
"description": "Confirm the user has permission to process any personal data in the inputs."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Upload file to private library",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": true
}
}List your reusable image, video, voice, music and sound effect assets.
No parameters.
{
"name": "assets_list",
"title": "List my assets",
"description": "List your reusable image, video, voice, music and sound effect assets.",
"inputSchema": {
"type": "object",
"required": [],
"properties": {},
"additionalProperties": false
},
"annotations": {
"title": "List my assets",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}List work history for one studio.
| Parameter | Type | Description |
|---|---|---|
| service * | string | Studio that owns the source, or whose details or requests you want to read. |
{
"name": "requests_list",
"title": "List studio requests",
"description": "List work history for one studio.",
"inputSchema": {
"type": "object",
"required": [
"service"
],
"properties": {
"service": {
"enum": [
"image",
"edit",
"upscale",
"voice",
"music",
"sfx",
"video",
"movie",
"media"
],
"type": "string",
"description": "Studio that owns the source, or whose details or requests you want to read."
}
},
"additionalProperties": false
},
"annotations": {
"title": "List studio requests",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Read request status, progress, results and analysis text.
| Parameter | Type | Description |
|---|---|---|
| service * | string | Studio that owns the source, or whose details or requests you want to read. |
| request_id * | string | UUID of the studio request. |
{
"name": "request_get",
"title": "Get request status",
"description": "Read request status, progress, results and analysis text.",
"inputSchema": {
"type": "object",
"required": [
"service",
"request_id"
],
"properties": {
"service": {
"enum": [
"image",
"edit",
"upscale",
"voice",
"music",
"sfx",
"video",
"movie",
"media"
],
"type": "string",
"description": "Studio that owns the source, or whose details or requests you want to read."
},
"request_id": {
"type": "string",
"format": "uuid",
"description": "UUID of the studio request."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Get request status",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Read a completed output in base64 chunks (up to 1 MiB). Follow next_offset until null. Binary files remain private.
| Parameter | Type | Description |
|---|---|---|
| offset | integer | Byte offset for the next result chunk, starting at zero. |
| service * | string | Studio that owns the source, or whose details or requests you want to read. |
| variant | string | Named output variant to read; available variants depend on the studio and request. |
| request_id * | string | UUID of the studio request. |
{
"name": "result_read",
"title": "Read result file",
"description": "Read a completed output in base64 chunks (up to 1 MiB). Follow next_offset until null. Binary files remain private.",
"inputSchema": {
"type": "object",
"required": [
"service",
"request_id"
],
"properties": {
"offset": {
"type": "integer",
"minimum": 0,
"description": "Byte offset for the next result chunk, starting at zero."
},
"service": {
"enum": [
"image",
"edit",
"upscale",
"voice",
"music",
"sfx",
"video",
"movie",
"media"
],
"type": "string",
"description": "Studio that owns the source, or whose details or requests you want to read."
},
"variant": {
"enum": [
"result",
"base",
"nsfw",
"upscaled",
"4k",
"first",
"last",
"metadata",
"project",
"log"
],
"type": "string",
"description": "Named output variant to read; available variants depend on the studio and request."
},
"request_id": {
"type": "string",
"format": "uuid",
"description": "UUID of the studio request."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Read result file",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Validate a Blender bpy Python program, output contract and CPU limits without execution or GPU jobs. Read blender_programming from capabilities_get. Owned assets are checked by movie_plan_validate.
| Parameter | Type | Description |
|---|---|---|
| clips | array | Ordered visual clips with their durations, sources and optional trims. |
| music | any | Optional background music track with placement on the movie timeline. |
| title | string | Display title for the plan or movie. |
| voice | any | Optional narration track with placement on the movie timeline. |
| aspect | string | Aspect ratio category for the movie output. |
| version | integer | Movie timeline schema version. |
| duration | any | Target movie duration in seconds. |
| graphics | array | Timed on-screen elements such as titles and captions. |
| sync_plan | any | Submit before Voice finishes. Bind a future voice asset; the server reads its output timing and resolves sentence/scene placement automatically after completion. |
| audio_layers | array | Optional audio tracks for voice, music and sound effects, with timing and gain. |
| blender_program | any | Optional Blender scene or timeline edit with its assets and output settings. |
| acknowledge_rights | boolean | Confirm the user has the rights or permissions to use the source material. |
| acknowledge_privacy | boolean | Confirm the user has permission to process any personal data in the inputs. |
{
"name": "blender_program_validate",
"title": "Validate Blender program",
"description": "Validate a Blender bpy Python program, output contract and CPU limits without execution or GPU jobs. Read blender_programming from capabilities_get. Owned assets are checked by movie_plan_validate.",
"inputSchema": {
"type": "object",
"$defs": {
"Cue": {
"type": "object",
"title": "Cue",
"required": [
"start",
"end",
"text"
],
"properties": {
"end": {
"type": "number",
"title": "End",
"maximum": 180,
"description": "Seconds in the original audio, exclusive.",
"exclusiveMinimum": 0
},
"text": {
"type": "string",
"title": "Text",
"maxLength": 160,
"minLength": 1
},
"start": {
"type": "number",
"title": "Start",
"minimum": 0,
"description": "Seconds in the original audio, inclusive.",
"exclusiveMaximum": 180
}
},
"additionalProperties": false
},
"Clip": {
"type": "object",
"title": "Clip",
"required": [
"asset",
"duration"
],
"properties": {
"trim": {
"type": "number",
"title": "Trim",
"default": 0,
"maximum": 180,
"minimum": 0
},
"asset": {
"$ref": "#/$defs/AssetRef"
},
"volume": {
"type": "number",
"title": "Volume",
"default": 1,
"maximum": 2,
"minimum": 0
},
"duration": {
"type": "number",
"title": "Duration",
"maximum": 120,
"minimum": 0.25
},
"hold_last_frame": {
"type": "boolean",
"title": "Hold Last Frame",
"default": false,
"description": "Allow extending a video by holding its final frame and padding audio with silence."
}
},
"additionalProperties": false
},
"AssetRef": {
"type": "object",
"title": "AssetRef",
"required": [
"service",
"id"
],
"properties": {
"id": {
"type": "string",
"title": "Id",
"format": "uuid"
},
"service": {
"enum": [
"upload",
"video",
"movie",
"image",
"edit",
"upscale",
"voice",
"music",
"sfx"
],
"type": "string",
"title": "Service"
},
"variant": {
"enum": [
"base",
"nsfw",
"result",
"upscaled",
"4k",
"first",
"last"
],
"type": "string",
"title": "Variant",
"default": "result"
}
},
"additionalProperties": false
},
"GainPoint": {
"type": "object",
"title": "GainPoint",
"required": [
"time",
"volume"
],
"properties": {
"time": {
"type": "number",
"title": "Time",
"maximum": 120,
"minimum": 0,
"description": "Seconds relative to the layer start."
},
"volume": {
"type": "number",
"title": "Volume",
"maximum": 2,
"minimum": 0
}
},
"additionalProperties": false
},
"AudioLayer": {
"type": "object",
"title": "AudioLayer",
"required": [
"asset",
"id",
"source_duration",
"duration"
],
"properties": {
"id": {
"type": "string",
"title": "Id",
"pattern": "^[A-Za-z0-9_-]{1,64}$"
},
"cues": {
"type": "array",
"items": {
"$ref": "#/$defs/Cue"
},
"title": "Cues",
"maxItems": 120,
"description": "Verified/imported source timestamps; never infer speech alignment from text length."
},
"duck": {
"type": "boolean",
"title": "Duck",
"default": false
},
"fade": {
"type": "number",
"title": "Fade",
"default": 1,
"maximum": 5,
"minimum": 0
},
"loop": {
"type": "boolean",
"title": "Loop",
"default": false
},
"name": {
"type": "string",
"title": "Name",
"default": "오디오 레이어",
"maxLength": 120
},
"role": {
"enum": [
"voice",
"music",
"sfx"
],
"type": "string",
"title": "Role",
"default": "voice"
},
"trim": {
"type": "number",
"title": "Trim",
"default": 0,
"maximum": 180,
"minimum": 0
},
"asset": {
"$ref": "#/$defs/AssetRef",
"description": "Audio-only source: Voice, Music, SFX result or an uploaded audio file. Video/Movie assets belong in clips; use clips.volume for their sound or privately upload extracted audio. Uploaded media kind is checked at dispatch."
},
"muted": {
"type": "boolean",
"title": "Muted",
"default": false
},
"offset": {
"type": "number",
"title": "Offset",
"default": 0,
"minimum": 0,
"exclusiveMaximum": 120
},
"volume": {
"type": "number",
"title": "Volume",
"default": 1,
"maximum": 2,
"minimum": 0
},
"captions": {
"type": "boolean",
"title": "Captions",
"default": false
},
"duration": {
"type": "number",
"title": "Duration",
"maximum": 120,
"minimum": 0.03333333333333333,
"description": "Length on the movie timeline; no speed change. Non-looping layers cannot exceed source_duration."
},
"envelope": {
"type": "array",
"items": {
"$ref": "#/$defs/GainPoint"
},
"title": "Envelope",
"maxItems": 32
},
"clip_index": {
"anyOf": [
{
"type": "integer",
"maximum": 23,
"minimum": 0
},
{
"type": "null"
}
],
"title": "Clip Index",
"default": null,
"description": "Optional scene anchor. offset is relative to this scene start; resolved after measured narration changes scene lengths. Omit for absolute movie seconds."
},
"caption_preset": {
"enum": [
"fade",
"slide_up",
"pop",
"typewriter"
],
"type": "string",
"title": "Caption Preset",
"default": "fade"
},
"preserve_source": {
"type": "boolean",
"title": "Preserve Source",
"default": false,
"description": "Protect a complete narration source: trim=0, no loop/fade/envelope/mute, full measured file duration. Timeline Blender programs must preserve this audio and resolved visual boundaries."
},
"source_duration": {
"type": "number",
"title": "Source Duration",
"maximum": 180,
"description": "Length of the selected source region, not the entire file. Loop repeats only this region.",
"exclusiveMinimum": 0
}
},
"additionalProperties": false
},
"AudioTrack": {
"type": "object",
"title": "AudioTrack",
"required": [
"asset"
],
"properties": {
"fade": {
"type": "number",
"title": "Fade",
"default": 1,
"maximum": 5,
"minimum": 0
},
"loop": {
"type": "boolean",
"title": "Loop",
"default": false
},
"trim": {
"type": "number",
"title": "Trim",
"default": 0,
"maximum": 180,
"minimum": 0
},
"asset": {
"$ref": "#/$defs/AssetRef",
"description": "Audio-only source: Voice, Music, SFX result or an uploaded audio file. Video/Movie assets belong in clips; use clips.volume for their sound or privately upload extracted audio. Uploaded media kind is checked at dispatch."
},
"offset": {
"type": "number",
"title": "Offset",
"default": 0,
"minimum": 0,
"exclusiveMaximum": 120
},
"volume": {
"type": "number",
"title": "Volume",
"default": 1,
"maximum": 2,
"minimum": 0
}
},
"additionalProperties": false
},
"ProgramAsset": {
"type": "object",
"title": "ProgramAsset",
"required": [
"name",
"asset"
],
"properties": {
"name": {
"type": "string",
"title": "Name",
"pattern": "^[A-Za-z][A-Za-z0-9_-]{0,39}$"
},
"asset": {
"$ref": "#/$defs/AssetRef"
}
},
"additionalProperties": false
},
"MotionGraphic": {
"type": "object",
"title": "MotionGraphic",
"required": [
"id",
"text",
"start",
"end"
],
"properties": {
"x": {
"type": "number",
"title": "X",
"default": 0.5,
"maximum": 0.9,
"minimum": 0.1,
"description": "Normalized horizontal center, left to right."
},
"y": {
"type": "number",
"title": "Y",
"default": 0.5,
"maximum": 0.9,
"minimum": 0.1,
"description": "Normalized vertical center, top to bottom."
},
"id": {
"type": "string",
"title": "Id",
"pattern": "^[A-Za-z0-9_-]{1,64}$"
},
"end": {
"type": "number",
"title": "End",
"maximum": 120,
"exclusiveMinimum": 0
},
"size": {
"type": "number",
"title": "Size",
"default": 0.065,
"maximum": 0.15,
"minimum": 0.02,
"description": "Font size as fraction of output height."
},
"text": {
"type": "string",
"title": "Text",
"maxLength": 160,
"minLength": 1
},
"color": {
"type": "string",
"title": "Color",
"default": "#ffffff",
"pattern": "^#[0-9a-fA-F]{6}$"
},
"start": {
"type": "number",
"title": "Start",
"minimum": 0,
"exclusiveMaximum": 120
},
"preset": {
"enum": [
"fade",
"slide_up",
"pop",
"typewriter"
],
"type": "string",
"title": "Preset",
"default": "slide_up"
},
"animation": {
"type": "number",
"title": "Animation",
"default": 0.3,
"maximum": 2,
"minimum": 0
},
"clip_index": {
"anyOf": [
{
"type": "integer",
"maximum": 23,
"minimum": 0
},
{
"type": "null"
}
],
"title": "Clip Index",
"default": null,
"description": "When supplied, start/end are relative to this scene; resolved after narration timing."
},
"follow_clip": {
"type": "boolean",
"title": "Follow Clip",
"default": false,
"description": "With clip_index, extend the title until the actual end of that scene."
}
},
"additionalProperties": false
},
"NarrationSync": {
"type": "object",
"title": "NarrationSync",
"required": [
"segments"
],
"properties": {
"mode": {
"enum": [
"segments",
"continuous",
"scene_sources"
],
"type": "string",
"title": "Mode",
"default": "segments",
"description": "Audio policy. continuous: play ONE complete source once, preserve original pauses, cut visuals at measured sentence starts; requires all sentence IDs in order. scene_sources: each segment plays ITS OWN complete source file (scene Voice clone) once from sample 0 to the measured file end, never trimmed to ASR cue ends; requires all of that source's sentence IDs. segments: legacy SELECTED-REGION edit that trims each scene to ASR sentence spans with 20ms fades; it does not promise full-source coverage. Omitted mode means segments for backward compatibility; new plans should always state the mode."
},
"asset": {
"anyOf": [
{
"$ref": "#/$defs/AssetRef"
},
{
"type": "null"
}
],
"default": null,
"description": "Default narration for segments without their own asset. May be omitted when every segment has an asset."
},
"segments": {
"type": "array",
"items": {
"$ref": "#/$defs/NarrationSegment"
},
"title": "Segments",
"maxItems": 24,
"minItems": 1
},
"end_guard": {
"type": "number",
"title": "End Guard",
"default": 1,
"maximum": 3,
"minimum": 1,
"description": "continuous and scene_sources: mandatory silence after the actual complete file end of the final narrated scene; never shrunk. ASR sentence end is NOT an audio trim boundary."
},
"visual_fit": {
"enum": [
"hold",
"strict"
],
"type": "string",
"title": "Visual Fit",
"default": "hold"
},
"start_padding": {
"type": "number",
"title": "Start Padding",
"default": 0.15,
"maximum": 3,
"minimum": 0,
"description": "continuous only: silence before the whole source, after any unbound intro clips."
},
"padding_policy": {
"enum": [
"preserve",
"shrink"
],
"type": "string",
"title": "Padding Policy",
"default": "preserve",
"description": "preserve keeps requested lead/tail. With fixed duration, shrink may proportionally reduce only lead/tail to fit; speech is never cut or sped up. For exact 15s shorts use fixed + shrink. If speech alone exceeds the budget, shorten the script or select fewer sentences."
},
"duration_policy": {
"enum": [
"fixed",
"expand"
],
"type": "string",
"title": "Duration Policy",
"default": "expand",
"description": "expand chooses the next allowed duration that fits actual speech, up to 120 seconds. fixed stops if speech exceeds the selected duration."
}
},
"additionalProperties": false
},
"BlenderProgram": {
"type": "object",
"title": "BlenderProgram",
"required": [
"summary",
"python"
],
"properties": {
"mode": {
"enum": [
"scene",
"timeline"
],
"type": "string",
"title": "Mode",
"default": "scene",
"description": "scene builds a new 3D/VSE/compositor scene; timeline edits the prepared Movie VSE project including its timed audio and text."
},
"assets": {
"type": "array",
"items": {
"$ref": "#/$defs/ProgramAsset"
},
"title": "Assets",
"maxItems": 24
},
"python": {
"type": "string",
"title": "Python",
"maxLength": 100000,
"minLength": 1,
"description": "Blender 5.2.1 bpy Python. Runs in an isolated CPU container, never the web/worker process. Build or edit bpy.context.scene; server renders it. HIRU contains resolved timeline, assets and output dimensions. No network, credentials, external packages or GPU."
},
"purpose": {
"enum": [
"movie",
"video_reference"
],
"type": "string",
"title": "Purpose",
"default": "movie"
},
"summary": {
"type": "string",
"title": "Summary",
"maxLength": 2000,
"minLength": 10,
"description": "Human-readable creative intent, including camera, animation, compositing and audio changes. Shown in Movie history."
},
"reference_kind": {
"enum": [
"rgb",
"depth",
"canny",
"pose"
],
"type": "string",
"title": "Reference Kind",
"default": "rgb",
"description": "Representation rendered by this script. Non-RGB control maps require a matching verified LTX IC-LoRA path."
},
"timeline_layout": {
"enum": [
"joined",
"clips"
],
"type": "string",
"title": "Timeline Layout",
"default": "joined",
"description": "joined preserves the legacy single MOVIE strip. clips exposes each original photo as IMAGE and each prepared video as MOVIE, named clip-0, clip-1, etc. Use clips for per-photo zoom/pan; audio and captions stay separate."
},
"required_transform_clips": {
"type": "array",
"items": {
"type": "integer"
},
"title": "Required Transform Clips",
"maxItems": 24,
"description": "With timeline_layout=clips, clip indices whose scale/offset/rotation must change during their own frame interval. A missing, muted, or static target fails before rendering; use for promised photo zoom/pan. This is transform validation, not proof of arbitrary visual effects."
}
},
"additionalProperties": false
},
"NarrationSegment": {
"type": "object",
"title": "NarrationSegment",
"required": [
"clip_index",
"sentence_ids"
],
"properties": {
"lead": {
"type": "number",
"title": "Lead",
"default": 0.15,
"maximum": 3,
"minimum": 0
},
"tail": {
"type": "number",
"title": "Tail",
"default": 0.25,
"maximum": 3,
"minimum": 0
},
"asset": {
"anyOf": [
{
"$ref": "#/$defs/AssetRef"
},
{
"type": "null"
}
],
"default": null,
"description": "Optional per-scene narration. Overrides sync_plan.asset; sentence IDs belong to this source. Bind each scene clone output asset here."
},
"captions": {
"type": "boolean",
"title": "Captions",
"default": true
},
"clip_index": {
"type": "integer",
"title": "Clip Index",
"maximum": 23,
"minimum": 0
},
"sentence_ids": {
"type": "array",
"items": {
"type": "string"
},
"title": "Sentence Ids",
"maxItems": 120,
"minItems": 1,
"description": "Stable source transcript IDs s0, s1, ...; one sentence per line is recommended."
},
"caption_preset": {
"enum": [
"fade",
"slide_up",
"pop",
"typewriter"
],
"type": "string",
"title": "Caption Preset",
"default": "fade"
}
},
"additionalProperties": false
}
},
"title": "Timeline",
"properties": {
"clips": {
"type": "array",
"items": {
"$ref": "#/$defs/Clip"
},
"title": "Clips",
"maxItems": 24,
"description": "Ordered visual clips with their durations, sources and optional trims."
},
"music": {
"anyOf": [
{
"$ref": "#/$defs/AudioTrack"
},
{
"type": "null"
}
],
"default": null,
"description": "Optional background music track with placement on the movie timeline."
},
"title": {
"type": "string",
"title": "Title",
"default": "나의 짧은 영화",
"maxLength": 120,
"minLength": 1,
"description": "Display title for the plan or movie."
},
"voice": {
"anyOf": [
{
"$ref": "#/$defs/AudioTrack"
},
{
"type": "null"
}
],
"default": null,
"description": "Optional narration track with placement on the movie timeline."
},
"aspect": {
"enum": [
"landscape",
"portrait",
"square"
],
"type": "string",
"title": "Aspect",
"default": "landscape",
"description": "Aspect ratio category for the movie output."
},
"version": {
"enum": [
1,
2
],
"type": "integer",
"title": "Version",
"default": 2,
"description": "Movie timeline schema version."
},
"duration": {
"anyOf": [
{
"enum": [
2,
3,
4,
5
],
"type": "integer"
},
{
"type": "integer",
"maximum": 120,
"minimum": 10
}
],
"title": "Duration",
"default": 15,
"description": "Target movie duration in seconds."
},
"graphics": {
"type": "array",
"items": {
"$ref": "#/$defs/MotionGraphic"
},
"title": "Graphics",
"maxItems": 64,
"description": "Timed on-screen elements such as titles and captions."
},
"sync_plan": {
"anyOf": [
{
"$ref": "#/$defs/NarrationSync"
},
{
"type": "null"
}
],
"default": null,
"description": "Submit before Voice finishes. Bind a future voice asset; the server reads its output timing and resolves sentence/scene placement automatically after completion."
},
"audio_layers": {
"type": "array",
"items": {
"$ref": "#/$defs/AudioLayer"
},
"title": "Audio Layers",
"maxItems": 32,
"description": "Optional audio tracks for voice, music and sound effects, with timing and gain."
},
"blender_program": {
"anyOf": [
{
"$ref": "#/$defs/BlenderProgram"
},
{
"type": "null"
}
],
"default": null,
"description": "Optional Blender scene or timeline edit with its assets and output settings."
},
"acknowledge_rights": {
"type": "boolean",
"title": "Acknowledge Rights",
"default": false,
"description": "Confirm the user has the rights or permissions to use the source material."
},
"acknowledge_privacy": {
"type": "boolean",
"title": "Acknowledge Privacy",
"default": false,
"description": "Confirm the user has permission to process any personal data in the inputs."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Validate Blender program",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Get stable sentence IDs from the exact planned Voice text BEFORE generation. Use these IDs in sync_plan.segments so a final edit plan can be submitted now and resolved automatically when narration completes.
| Parameter | Type | Description |
|---|---|---|
| text * | string | Exact narration text planned for the Voice request. |
{
"name": "movie_script_analyze",
"title": "Analyze narration script",
"description": "Get stable sentence IDs from the exact planned Voice text BEFORE generation. Use these IDs in sync_plan.segments so a final edit plan can be submitted now and resolved automatically when narration completes.",
"inputSchema": {
"type": "object",
"title": "ScriptAnalysis",
"required": [
"text"
],
"properties": {
"text": {
"type": "string",
"title": "Text",
"maxLength": 8000,
"minLength": 1,
"description": "Exact narration text planned for the Voice request."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Analyze narration script",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Inspect actual audio duration, bounded waveform and imported subtitle cues without generation or GPU jobs.
| Parameter | Type | Description |
|---|---|---|
| id * | string | UUID of the source asset. |
| service * | string | Studio that owns the source, or whose details or requests you want to read. |
| variant | string | Named output variant to read; available variants depend on the studio and request. |
{
"name": "movie_asset_metadata",
"title": "Inspect audio asset",
"description": "Inspect actual audio duration, bounded waveform and imported subtitle cues without generation or GPU jobs.",
"inputSchema": {
"type": "object",
"title": "AssetRef",
"required": [
"service",
"id"
],
"properties": {
"id": {
"type": "string",
"title": "Id",
"format": "uuid",
"description": "UUID of the source asset."
},
"service": {
"enum": [
"upload",
"video",
"movie",
"image",
"edit",
"upscale",
"voice",
"music",
"sfx"
],
"type": "string",
"title": "Service",
"description": "Studio that owns the source, or whose details or requests you want to read."
},
"variant": {
"enum": [
"base",
"nsfw",
"result",
"upscaled",
"4k",
"first",
"last"
],
"type": "string",
"title": "Variant",
"default": "result",
"description": "Named output variant to read; available variants depend on the studio and request."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Inspect audio asset",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Validate an edit against owned real source durations, mapped narration cues and layer limits. No render, email, quota or GPU job. Use after capabilities_get and before submission.
| Parameter | Type | Description |
|---|---|---|
| clips | array | Ordered visual clips with their durations, sources and optional trims. |
| music | any | Optional background music track with placement on the movie timeline. |
| title | string | Display title for the plan or movie. |
| voice | any | Optional narration track with placement on the movie timeline. |
| aspect | string | Aspect ratio category for the movie output. |
| version | integer | Movie timeline schema version. |
| duration | any | Target movie duration in seconds. |
| graphics | array | Timed on-screen elements such as titles and captions. |
| sync_plan | any | Submit before Voice finishes. Bind a future voice asset; the server reads its output timing and resolves sentence/scene placement automatically after completion. |
| audio_layers | array | Optional audio tracks for voice, music and sound effects, with timing and gain. |
| blender_program | any | Optional Blender scene or timeline edit with its assets and output settings. |
| acknowledge_rights | boolean | Confirm the user has the rights or permissions to use the source material. |
| acknowledge_privacy | boolean | Confirm the user has permission to process any personal data in the inputs. |
{
"name": "movie_plan_validate",
"title": "Validate movie edit",
"description": "Validate an edit against owned real source durations, mapped narration cues and layer limits. No render, email, quota or GPU job. Use after capabilities_get and before submission.",
"inputSchema": {
"type": "object",
"$defs": {
"Cue": {
"type": "object",
"title": "Cue",
"required": [
"start",
"end",
"text"
],
"properties": {
"end": {
"type": "number",
"title": "End",
"maximum": 180,
"description": "Seconds in the original audio, exclusive.",
"exclusiveMinimum": 0
},
"text": {
"type": "string",
"title": "Text",
"maxLength": 160,
"minLength": 1
},
"start": {
"type": "number",
"title": "Start",
"minimum": 0,
"description": "Seconds in the original audio, inclusive.",
"exclusiveMaximum": 180
}
},
"additionalProperties": false
},
"Clip": {
"type": "object",
"title": "Clip",
"required": [
"asset",
"duration"
],
"properties": {
"trim": {
"type": "number",
"title": "Trim",
"default": 0,
"maximum": 180,
"minimum": 0
},
"asset": {
"$ref": "#/$defs/AssetRef"
},
"volume": {
"type": "number",
"title": "Volume",
"default": 1,
"maximum": 2,
"minimum": 0
},
"duration": {
"type": "number",
"title": "Duration",
"maximum": 120,
"minimum": 0.25
},
"hold_last_frame": {
"type": "boolean",
"title": "Hold Last Frame",
"default": false,
"description": "Allow extending a video by holding its final frame and padding audio with silence."
}
},
"additionalProperties": false
},
"AssetRef": {
"type": "object",
"title": "AssetRef",
"required": [
"service",
"id"
],
"properties": {
"id": {
"type": "string",
"title": "Id",
"format": "uuid"
},
"service": {
"enum": [
"upload",
"video",
"movie",
"image",
"edit",
"upscale",
"voice",
"music",
"sfx"
],
"type": "string",
"title": "Service"
},
"variant": {
"enum": [
"base",
"nsfw",
"result",
"upscaled",
"4k",
"first",
"last"
],
"type": "string",
"title": "Variant",
"default": "result"
}
},
"additionalProperties": false
},
"GainPoint": {
"type": "object",
"title": "GainPoint",
"required": [
"time",
"volume"
],
"properties": {
"time": {
"type": "number",
"title": "Time",
"maximum": 120,
"minimum": 0,
"description": "Seconds relative to the layer start."
},
"volume": {
"type": "number",
"title": "Volume",
"maximum": 2,
"minimum": 0
}
},
"additionalProperties": false
},
"AudioLayer": {
"type": "object",
"title": "AudioLayer",
"required": [
"asset",
"id",
"source_duration",
"duration"
],
"properties": {
"id": {
"type": "string",
"title": "Id",
"pattern": "^[A-Za-z0-9_-]{1,64}$"
},
"cues": {
"type": "array",
"items": {
"$ref": "#/$defs/Cue"
},
"title": "Cues",
"maxItems": 120,
"description": "Verified/imported source timestamps; never infer speech alignment from text length."
},
"duck": {
"type": "boolean",
"title": "Duck",
"default": false
},
"fade": {
"type": "number",
"title": "Fade",
"default": 1,
"maximum": 5,
"minimum": 0
},
"loop": {
"type": "boolean",
"title": "Loop",
"default": false
},
"name": {
"type": "string",
"title": "Name",
"default": "오디오 레이어",
"maxLength": 120
},
"role": {
"enum": [
"voice",
"music",
"sfx"
],
"type": "string",
"title": "Role",
"default": "voice"
},
"trim": {
"type": "number",
"title": "Trim",
"default": 0,
"maximum": 180,
"minimum": 0
},
"asset": {
"$ref": "#/$defs/AssetRef",
"description": "Audio-only source: Voice, Music, SFX result or an uploaded audio file. Video/Movie assets belong in clips; use clips.volume for their sound or privately upload extracted audio. Uploaded media kind is checked at dispatch."
},
"muted": {
"type": "boolean",
"title": "Muted",
"default": false
},
"offset": {
"type": "number",
"title": "Offset",
"default": 0,
"minimum": 0,
"exclusiveMaximum": 120
},
"volume": {
"type": "number",
"title": "Volume",
"default": 1,
"maximum": 2,
"minimum": 0
},
"captions": {
"type": "boolean",
"title": "Captions",
"default": false
},
"duration": {
"type": "number",
"title": "Duration",
"maximum": 120,
"minimum": 0.03333333333333333,
"description": "Length on the movie timeline; no speed change. Non-looping layers cannot exceed source_duration."
},
"envelope": {
"type": "array",
"items": {
"$ref": "#/$defs/GainPoint"
},
"title": "Envelope",
"maxItems": 32
},
"clip_index": {
"anyOf": [
{
"type": "integer",
"maximum": 23,
"minimum": 0
},
{
"type": "null"
}
],
"title": "Clip Index",
"default": null,
"description": "Optional scene anchor. offset is relative to this scene start; resolved after measured narration changes scene lengths. Omit for absolute movie seconds."
},
"caption_preset": {
"enum": [
"fade",
"slide_up",
"pop",
"typewriter"
],
"type": "string",
"title": "Caption Preset",
"default": "fade"
},
"preserve_source": {
"type": "boolean",
"title": "Preserve Source",
"default": false,
"description": "Protect a complete narration source: trim=0, no loop/fade/envelope/mute, full measured file duration. Timeline Blender programs must preserve this audio and resolved visual boundaries."
},
"source_duration": {
"type": "number",
"title": "Source Duration",
"maximum": 180,
"description": "Length of the selected source region, not the entire file. Loop repeats only this region.",
"exclusiveMinimum": 0
}
},
"additionalProperties": false
},
"AudioTrack": {
"type": "object",
"title": "AudioTrack",
"required": [
"asset"
],
"properties": {
"fade": {
"type": "number",
"title": "Fade",
"default": 1,
"maximum": 5,
"minimum": 0
},
"loop": {
"type": "boolean",
"title": "Loop",
"default": false
},
"trim": {
"type": "number",
"title": "Trim",
"default": 0,
"maximum": 180,
"minimum": 0
},
"asset": {
"$ref": "#/$defs/AssetRef",
"description": "Audio-only source: Voice, Music, SFX result or an uploaded audio file. Video/Movie assets belong in clips; use clips.volume for their sound or privately upload extracted audio. Uploaded media kind is checked at dispatch."
},
"offset": {
"type": "number",
"title": "Offset",
"default": 0,
"minimum": 0,
"exclusiveMaximum": 120
},
"volume": {
"type": "number",
"title": "Volume",
"default": 1,
"maximum": 2,
"minimum": 0
}
},
"additionalProperties": false
},
"ProgramAsset": {
"type": "object",
"title": "ProgramAsset",
"required": [
"name",
"asset"
],
"properties": {
"name": {
"type": "string",
"title": "Name",
"pattern": "^[A-Za-z][A-Za-z0-9_-]{0,39}$"
},
"asset": {
"$ref": "#/$defs/AssetRef"
}
},
"additionalProperties": false
},
"MotionGraphic": {
"type": "object",
"title": "MotionGraphic",
"required": [
"id",
"text",
"start",
"end"
],
"properties": {
"x": {
"type": "number",
"title": "X",
"default": 0.5,
"maximum": 0.9,
"minimum": 0.1,
"description": "Normalized horizontal center, left to right."
},
"y": {
"type": "number",
"title": "Y",
"default": 0.5,
"maximum": 0.9,
"minimum": 0.1,
"description": "Normalized vertical center, top to bottom."
},
"id": {
"type": "string",
"title": "Id",
"pattern": "^[A-Za-z0-9_-]{1,64}$"
},
"end": {
"type": "number",
"title": "End",
"maximum": 120,
"exclusiveMinimum": 0
},
"size": {
"type": "number",
"title": "Size",
"default": 0.065,
"maximum": 0.15,
"minimum": 0.02,
"description": "Font size as fraction of output height."
},
"text": {
"type": "string",
"title": "Text",
"maxLength": 160,
"minLength": 1
},
"color": {
"type": "string",
"title": "Color",
"default": "#ffffff",
"pattern": "^#[0-9a-fA-F]{6}$"
},
"start": {
"type": "number",
"title": "Start",
"minimum": 0,
"exclusiveMaximum": 120
},
"preset": {
"enum": [
"fade",
"slide_up",
"pop",
"typewriter"
],
"type": "string",
"title": "Preset",
"default": "slide_up"
},
"animation": {
"type": "number",
"title": "Animation",
"default": 0.3,
"maximum": 2,
"minimum": 0
},
"clip_index": {
"anyOf": [
{
"type": "integer",
"maximum": 23,
"minimum": 0
},
{
"type": "null"
}
],
"title": "Clip Index",
"default": null,
"description": "When supplied, start/end are relative to this scene; resolved after narration timing."
},
"follow_clip": {
"type": "boolean",
"title": "Follow Clip",
"default": false,
"description": "With clip_index, extend the title until the actual end of that scene."
}
},
"additionalProperties": false
},
"NarrationSync": {
"type": "object",
"title": "NarrationSync",
"required": [
"segments"
],
"properties": {
"mode": {
"enum": [
"segments",
"continuous",
"scene_sources"
],
"type": "string",
"title": "Mode",
"default": "segments",
"description": "Audio policy. continuous: play ONE complete source once, preserve original pauses, cut visuals at measured sentence starts; requires all sentence IDs in order. scene_sources: each segment plays ITS OWN complete source file (scene Voice clone) once from sample 0 to the measured file end, never trimmed to ASR cue ends; requires all of that source's sentence IDs. segments: legacy SELECTED-REGION edit that trims each scene to ASR sentence spans with 20ms fades; it does not promise full-source coverage. Omitted mode means segments for backward compatibility; new plans should always state the mode."
},
"asset": {
"anyOf": [
{
"$ref": "#/$defs/AssetRef"
},
{
"type": "null"
}
],
"default": null,
"description": "Default narration for segments without their own asset. May be omitted when every segment has an asset."
},
"segments": {
"type": "array",
"items": {
"$ref": "#/$defs/NarrationSegment"
},
"title": "Segments",
"maxItems": 24,
"minItems": 1
},
"end_guard": {
"type": "number",
"title": "End Guard",
"default": 1,
"maximum": 3,
"minimum": 1,
"description": "continuous and scene_sources: mandatory silence after the actual complete file end of the final narrated scene; never shrunk. ASR sentence end is NOT an audio trim boundary."
},
"visual_fit": {
"enum": [
"hold",
"strict"
],
"type": "string",
"title": "Visual Fit",
"default": "hold"
},
"start_padding": {
"type": "number",
"title": "Start Padding",
"default": 0.15,
"maximum": 3,
"minimum": 0,
"description": "continuous only: silence before the whole source, after any unbound intro clips."
},
"padding_policy": {
"enum": [
"preserve",
"shrink"
],
"type": "string",
"title": "Padding Policy",
"default": "preserve",
"description": "preserve keeps requested lead/tail. With fixed duration, shrink may proportionally reduce only lead/tail to fit; speech is never cut or sped up. For exact 15s shorts use fixed + shrink. If speech alone exceeds the budget, shorten the script or select fewer sentences."
},
"duration_policy": {
"enum": [
"fixed",
"expand"
],
"type": "string",
"title": "Duration Policy",
"default": "expand",
"description": "expand chooses the next allowed duration that fits actual speech, up to 120 seconds. fixed stops if speech exceeds the selected duration."
}
},
"additionalProperties": false
},
"BlenderProgram": {
"type": "object",
"title": "BlenderProgram",
"required": [
"summary",
"python"
],
"properties": {
"mode": {
"enum": [
"scene",
"timeline"
],
"type": "string",
"title": "Mode",
"default": "scene",
"description": "scene builds a new 3D/VSE/compositor scene; timeline edits the prepared Movie VSE project including its timed audio and text."
},
"assets": {
"type": "array",
"items": {
"$ref": "#/$defs/ProgramAsset"
},
"title": "Assets",
"maxItems": 24
},
"python": {
"type": "string",
"title": "Python",
"maxLength": 100000,
"minLength": 1,
"description": "Blender 5.2.1 bpy Python. Runs in an isolated CPU container, never the web/worker process. Build or edit bpy.context.scene; server renders it. HIRU contains resolved timeline, assets and output dimensions. No network, credentials, external packages or GPU."
},
"purpose": {
"enum": [
"movie",
"video_reference"
],
"type": "string",
"title": "Purpose",
"default": "movie"
},
"summary": {
"type": "string",
"title": "Summary",
"maxLength": 2000,
"minLength": 10,
"description": "Human-readable creative intent, including camera, animation, compositing and audio changes. Shown in Movie history."
},
"reference_kind": {
"enum": [
"rgb",
"depth",
"canny",
"pose"
],
"type": "string",
"title": "Reference Kind",
"default": "rgb",
"description": "Representation rendered by this script. Non-RGB control maps require a matching verified LTX IC-LoRA path."
},
"timeline_layout": {
"enum": [
"joined",
"clips"
],
"type": "string",
"title": "Timeline Layout",
"default": "joined",
"description": "joined preserves the legacy single MOVIE strip. clips exposes each original photo as IMAGE and each prepared video as MOVIE, named clip-0, clip-1, etc. Use clips for per-photo zoom/pan; audio and captions stay separate."
},
"required_transform_clips": {
"type": "array",
"items": {
"type": "integer"
},
"title": "Required Transform Clips",
"maxItems": 24,
"description": "With timeline_layout=clips, clip indices whose scale/offset/rotation must change during their own frame interval. A missing, muted, or static target fails before rendering; use for promised photo zoom/pan. This is transform validation, not proof of arbitrary visual effects."
}
},
"additionalProperties": false
},
"NarrationSegment": {
"type": "object",
"title": "NarrationSegment",
"required": [
"clip_index",
"sentence_ids"
],
"properties": {
"lead": {
"type": "number",
"title": "Lead",
"default": 0.15,
"maximum": 3,
"minimum": 0
},
"tail": {
"type": "number",
"title": "Tail",
"default": 0.25,
"maximum": 3,
"minimum": 0
},
"asset": {
"anyOf": [
{
"$ref": "#/$defs/AssetRef"
},
{
"type": "null"
}
],
"default": null,
"description": "Optional per-scene narration. Overrides sync_plan.asset; sentence IDs belong to this source. Bind each scene clone output asset here."
},
"captions": {
"type": "boolean",
"title": "Captions",
"default": true
},
"clip_index": {
"type": "integer",
"title": "Clip Index",
"maximum": 23,
"minimum": 0
},
"sentence_ids": {
"type": "array",
"items": {
"type": "string"
},
"title": "Sentence Ids",
"maxItems": 120,
"minItems": 1,
"description": "Stable source transcript IDs s0, s1, ...; one sentence per line is recommended."
},
"caption_preset": {
"enum": [
"fade",
"slide_up",
"pop",
"typewriter"
],
"type": "string",
"title": "Caption Preset",
"default": "fade"
}
},
"additionalProperties": false
}
},
"title": "Timeline",
"properties": {
"clips": {
"type": "array",
"items": {
"$ref": "#/$defs/Clip"
},
"title": "Clips",
"maxItems": 24,
"description": "Ordered visual clips with their durations, sources and optional trims."
},
"music": {
"anyOf": [
{
"$ref": "#/$defs/AudioTrack"
},
{
"type": "null"
}
],
"default": null,
"description": "Optional background music track with placement on the movie timeline."
},
"title": {
"type": "string",
"title": "Title",
"default": "나의 짧은 영화",
"maxLength": 120,
"minLength": 1,
"description": "Display title for the plan or movie."
},
"voice": {
"anyOf": [
{
"$ref": "#/$defs/AudioTrack"
},
{
"type": "null"
}
],
"default": null,
"description": "Optional narration track with placement on the movie timeline."
},
"aspect": {
"enum": [
"landscape",
"portrait",
"square"
],
"type": "string",
"title": "Aspect",
"default": "landscape",
"description": "Aspect ratio category for the movie output."
},
"version": {
"enum": [
1,
2
],
"type": "integer",
"title": "Version",
"default": 2,
"description": "Movie timeline schema version."
},
"duration": {
"anyOf": [
{
"enum": [
2,
3,
4,
5
],
"type": "integer"
},
{
"type": "integer",
"maximum": 120,
"minimum": 10
}
],
"title": "Duration",
"default": 15,
"description": "Target movie duration in seconds."
},
"graphics": {
"type": "array",
"items": {
"$ref": "#/$defs/MotionGraphic"
},
"title": "Graphics",
"maxItems": 64,
"description": "Timed on-screen elements such as titles and captions."
},
"sync_plan": {
"anyOf": [
{
"$ref": "#/$defs/NarrationSync"
},
{
"type": "null"
}
],
"default": null,
"description": "Submit before Voice finishes. Bind a future voice asset; the server reads its output timing and resolves sentence/scene placement automatically after completion."
},
"audio_layers": {
"type": "array",
"items": {
"$ref": "#/$defs/AudioLayer"
},
"title": "Audio Layers",
"maxItems": 32,
"description": "Optional audio tracks for voice, music and sound effects, with timing and gain."
},
"blender_program": {
"anyOf": [
{
"$ref": "#/$defs/BlenderProgram"
},
{
"type": "null"
}
],
"default": null,
"description": "Optional Blender scene or timeline edit with its assets and output settings."
},
"acknowledge_rights": {
"type": "boolean",
"title": "Acknowledge Rights",
"default": false,
"description": "Confirm the user has the rights or permissions to use the source material."
},
"acknowledge_privacy": {
"type": "boolean",
"title": "Acknowledge Privacy",
"default": false,
"description": "Confirm the user has permission to process any personal data in the inputs."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Validate movie edit",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Save verified source timestamps for owned audio. Times are original-file seconds, not movie offsets. Never invent alignment from transcript length.
| Parameter | Type | Description |
|---|---|---|
| asset * | ref | Owned source audio asset to update. |
| timing * | ref | Verified cues in source-file seconds and their provenance. |
{
"name": "movie_asset_timing_set",
"title": "Save audio timing",
"description": "Save verified source timestamps for owned audio. Times are original-file seconds, not movie offsets. Never invent alignment from transcript length.",
"inputSchema": {
"type": "object",
"$defs": {
"Cue": {
"type": "object",
"title": "Cue",
"required": [
"start",
"end",
"text"
],
"properties": {
"end": {
"type": "number",
"title": "End",
"maximum": 180,
"description": "Seconds in the original audio, exclusive.",
"exclusiveMinimum": 0
},
"text": {
"type": "string",
"title": "Text",
"maxLength": 160,
"minLength": 1
},
"start": {
"type": "number",
"title": "Start",
"minimum": 0,
"description": "Seconds in the original audio, inclusive.",
"exclusiveMaximum": 180
}
},
"additionalProperties": false
},
"AssetRef": {
"type": "object",
"title": "AssetRef",
"required": [
"service",
"id"
],
"properties": {
"id": {
"type": "string",
"title": "Id",
"format": "uuid"
},
"service": {
"enum": [
"upload",
"video",
"movie",
"image",
"edit",
"upscale",
"voice",
"music",
"sfx"
],
"type": "string",
"title": "Service"
},
"variant": {
"enum": [
"base",
"nsfw",
"result",
"upscaled",
"4k",
"first",
"last"
],
"type": "string",
"title": "Variant",
"default": "result"
}
},
"additionalProperties": false
},
"AssetTiming": {
"type": "object",
"title": "AssetTiming",
"properties": {
"cues": {
"type": "array",
"items": {
"$ref": "#/$defs/Cue"
},
"title": "Cues",
"maxItems": 120
},
"provenance": {
"type": "string",
"title": "Provenance",
"default": "manual",
"pattern": "^(manual|imported_srt|imported_vtt|verified_alignment)$"
}
},
"additionalProperties": false
}
},
"title": "TimingUpdate",
"required": [
"asset",
"timing"
],
"properties": {
"asset": {
"$ref": "#/$defs/AssetRef",
"description": "Owned source audio asset to update."
},
"timing": {
"$ref": "#/$defs/AssetTiming",
"description": "Verified cues in source-file seconds and their provenance."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Save audio timing",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Start here: read the HiruAI workflow without installing skills. Get task-specific steps, consent and publishing guidance, recovery actions, or a full recipe with recipe_id even if this client cannot read MCP resources. Read-only; no job or quota.
| Parameter | Type | Description |
|---|---|---|
| topic | string | Read only the topic needed now; omitted means the first-use workflow. |
| recipe_id | string | Optionally include the full selected recipe, including for clients without resources/read. |
{
"name": "workflow_get",
"title": "Read workflow and recipes",
"description": "Start here: read the HiruAI workflow without installing skills. Get task-specific steps, consent and publishing guidance, recovery actions, or a full recipe with recipe_id even if this client cannot read MCP resources. Read-only; no job or quota.",
"inputSchema": {
"type": "object",
"properties": {
"topic": {
"enum": [
"start",
"image",
"shorts",
"narration",
"audio",
"video",
"media",
"recovery",
"publishing"
],
"type": "string",
"default": "start",
"description": "Read only the topic needed now; omitted means the first-use workflow."
},
"recipe_id": {
"enum": [
"shorts_full_width",
"narrated_broll",
"scene_voice_clones",
"speaking_character",
"video_continuity",
"exact_camera_path",
"kinetic_typography",
"music_led_montage"
],
"type": "string",
"description": "Optionally include the full selected recipe, including for clients without resources/read."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Read workflow and recipes",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Read live studio input schemas, supported outputs and planning rules before making a plan. Optional service returns only that studio and relevant details; omit it for the existing full catalog. Start with workflow_get for task routing. Creative hooks are a nudge, not a rejection rule.
| Parameter | Type | Description |
|---|---|---|
| service | string | Studio that owns the source, or whose details or requests you want to read. |
{
"name": "capabilities_get",
"title": "Read capabilities",
"description": "Read live studio input schemas, supported outputs and planning rules before making a plan. Optional service returns only that studio and relevant details; omit it for the existing full catalog. Start with workflow_get for task routing. Creative hooks are a nudge, not a rejection rule.",
"inputSchema": {
"type": "object",
"properties": {
"service": {
"enum": [
"image",
"edit",
"upscale",
"voice",
"music",
"sfx",
"video",
"movie",
"media"
],
"type": "string",
"description": "Studio that owns the source, or whose details or requests you want to read."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Read capabilities",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Validate a dependency-ordered plan without starting jobs or consuming studio quota. Read capabilities_get first. Declare the dopamine hook(s) built into the final work as dopamine_hooks; a creative plan without one stays valid and returns a dopamine_hook_missing suggestion; explicit user overrides are allowed.
| Parameter | Type | Description |
|---|---|---|
| goal * | string | Desired finished deliverable and user-provided requirements. |
| steps * | array | Production steps in dependency order, with studio arguments and result bindings. |
| title * | string | Display title for the plan or movie. |
| dopamine_hooks | array | Creative baseline: the dopamine hook(s) you built into the FINAL work. Declare at least one for any plan that creates image, voice, music, video or movie output, after reading dopamine_hooks_get. The server records each as a "[도파민 요소] …" line in goal, shown to the user on the dashboard. |
{
"name": "batch_validate",
"title": "Validate production plan",
"description": "Validate a dependency-ordered plan without starting jobs or consuming studio quota. Read capabilities_get first. Declare the dopamine hook(s) built into the final work as dopamine_hooks; a creative plan without one stays valid and returns a dopamine_hook_missing suggestion; explicit user overrides are allowed.",
"inputSchema": {
"type": "object",
"$defs": {
"Step": {
"type": "object",
"title": "Step",
"required": [
"key",
"title",
"service",
"arguments"
],
"properties": {
"key": {
"type": "string",
"title": "Key",
"pattern": "^[a-zA-Z][a-zA-Z0-9_-]{0,39}$"
},
"title": {
"type": "string",
"title": "Title",
"maxLength": 160,
"minLength": 1
},
"service": {
"enum": [
"image",
"edit",
"upscale",
"voice",
"music",
"sfx",
"video",
"movie",
"media"
],
"type": "string",
"title": "Service"
},
"bindings": {
"type": "array",
"items": {
"$ref": "#/$defs/Binding"
},
"title": "Bindings",
"maxItems": 72
},
"arguments": {
"type": "object",
"title": "Arguments",
"additionalProperties": true
},
"depends_on": {
"type": "array",
"items": {
"type": "string"
},
"title": "Depends On",
"maxItems": 32
}
},
"additionalProperties": false
},
"Binding": {
"type": "object",
"title": "Binding",
"required": [
"input",
"step"
],
"properties": {
"step": {
"type": "string",
"title": "Step",
"pattern": "^[a-zA-Z][a-zA-Z0-9_-]{0,39}$"
},
"input": {
"type": "string",
"title": "Input",
"pattern": "^[A-Za-z_][A-Za-z_0-9]*(\\.[A-Za-z_0-9]+)*$",
"maxLength": 160
},
"output": {
"enum": [
"id",
"asset"
],
"type": "string",
"title": "Output",
"default": "id"
},
"variant": {
"anyOf": [
{
"enum": [
"base",
"nsfw",
"result",
"upscaled",
"4k",
"first",
"last"
],
"type": "string"
},
{
"type": "null"
}
],
"title": "Variant",
"default": null
}
},
"additionalProperties": false
}
},
"title": "Plan",
"required": [
"title",
"goal",
"steps"
],
"properties": {
"goal": {
"type": "string",
"title": "Goal",
"maxLength": 4000,
"minLength": 1,
"description": "Desired finished deliverable and user-provided requirements."
},
"steps": {
"type": "array",
"items": {
"$ref": "#/$defs/Step"
},
"title": "Steps",
"maxItems": 32,
"minItems": 1,
"description": "Production steps in dependency order, with studio arguments and result bindings."
},
"title": {
"type": "string",
"title": "Title",
"maxLength": 160,
"minLength": 1,
"description": "Display title for the plan or movie."
},
"dopamine_hooks": {
"type": "array",
"items": {
"type": "object",
"required": [
"id",
"channel",
"where"
],
"properties": {
"id": {
"enum": [
"novelty",
"surprise_reward",
"uncertain_reward",
"small_wins",
"problem_solved",
"aha_moment",
"social_reward",
"winning",
"music_chills",
"exercise_high",
"awe_nature",
"scent_taste_preview",
"shopping_unboxing",
"cliffhanger",
"info_seeking",
"big_decision_closed",
"teamwork_sync",
"tension_release",
"altruism",
"caffeine_arousal",
"thirst_hunger_relief",
"attraction_anticipation",
"thermal_comfort",
"flow_state",
"humor_punchline",
"order_from_chaos",
"insider_secret",
"cuteness",
"planning_anticipation",
"outrage_justice",
"sugar_rush",
"procrastination_relief"
],
"type": "string"
},
"where": {
"type": "string",
"pattern": "^[^\\r\\n]+$",
"maxLength": 160,
"minLength": 4,
"description": "Where and how it appears in the finished work, in one line. Example: \"0~3초: 오븐 문이 열리며 김이 오르는 빵 클로즈업\"."
},
"channel": {
"enum": [
"visual",
"audio",
"situation"
],
"type": "string",
"description": "How the audience meets it: visual (시각), audio (음향) or situation (상황)."
}
},
"additionalProperties": false
},
"maxItems": 3,
"minItems": 1,
"description": "Creative baseline: the dopamine hook(s) you built into the FINAL work. Declare at least one for any plan that creates image, voice, music, video or movie output, after reading dopamine_hooks_get. The server records each as a \"[도파민 요소] …\" line in goal, shown to the user on the dashboard."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Validate production plan",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Save the entire plan once and immediately hand the returned dashboard URL to the user. Do NOT poll or wait for completion. Server handles dependencies, GPU waits, studio/email locks and quota retries. Default start=false lets the user review and start on the dashboard. start=true requires user authorization for automatic execution and every acknowledgement. A submitted plan continues after the chat closes or session expires. Send the same dopamine_hooks you validated; they are recorded in goal for the user to review.
| Parameter | Type | Description |
|---|---|---|
| goal * | string | Desired finished deliverable and user-provided requirements. |
| start | boolean | Start after submission when authorized; false saves a draft for review. |
| steps * | array | Production steps in dependency order, with studio arguments and result bindings. |
| title * | string | Display title for the plan or movie. |
| dopamine_hooks | array | Creative baseline: the dopamine hook(s) you built into the FINAL work. Declare at least one for any plan that creates image, voice, music, video or movie output, after reading dopamine_hooks_get. The server records each as a "[도파민 요소] …" line in goal, shown to the user on the dashboard. |
| idempotency_key * | string | Unique key for the intended operation. Reuse it on network retries. |
| acknowledge_execution | boolean | Confirm the user explicitly approved starting the production plan automatically. |
{
"name": "batch_submit",
"title": "Submit production plan",
"description": "Save the entire plan once and immediately hand the returned dashboard URL to the user. Do NOT poll or wait for completion. Server handles dependencies, GPU waits, studio/email locks and quota retries. Default start=false lets the user review and start on the dashboard. start=true requires user authorization for automatic execution and every acknowledgement. A submitted plan continues after the chat closes or session expires. Send the same dopamine_hooks you validated; they are recorded in goal for the user to review.",
"inputSchema": {
"type": "object",
"$defs": {
"Step": {
"type": "object",
"title": "Step",
"required": [
"key",
"title",
"service",
"arguments"
],
"properties": {
"key": {
"type": "string",
"title": "Key",
"pattern": "^[a-zA-Z][a-zA-Z0-9_-]{0,39}$"
},
"title": {
"type": "string",
"title": "Title",
"maxLength": 160,
"minLength": 1
},
"service": {
"enum": [
"image",
"edit",
"upscale",
"voice",
"music",
"sfx",
"video",
"movie",
"media"
],
"type": "string",
"title": "Service"
},
"bindings": {
"type": "array",
"items": {
"$ref": "#/$defs/Binding"
},
"title": "Bindings",
"maxItems": 72
},
"arguments": {
"type": "object",
"title": "Arguments",
"additionalProperties": true
},
"depends_on": {
"type": "array",
"items": {
"type": "string"
},
"title": "Depends On",
"maxItems": 32
}
},
"additionalProperties": false
},
"Binding": {
"type": "object",
"title": "Binding",
"required": [
"input",
"step"
],
"properties": {
"step": {
"type": "string",
"title": "Step",
"pattern": "^[a-zA-Z][a-zA-Z0-9_-]{0,39}$"
},
"input": {
"type": "string",
"title": "Input",
"pattern": "^[A-Za-z_][A-Za-z_0-9]*(\\.[A-Za-z_0-9]+)*$",
"maxLength": 160
},
"output": {
"enum": [
"id",
"asset"
],
"type": "string",
"title": "Output",
"default": "id"
},
"variant": {
"anyOf": [
{
"enum": [
"base",
"nsfw",
"result",
"upscaled",
"4k",
"first",
"last"
],
"type": "string"
},
{
"type": "null"
}
],
"title": "Variant",
"default": null
}
},
"additionalProperties": false
}
},
"title": "Submission",
"required": [
"title",
"goal",
"steps",
"idempotency_key"
],
"properties": {
"goal": {
"type": "string",
"title": "Goal",
"maxLength": 4000,
"minLength": 1,
"description": "Desired finished deliverable and user-provided requirements."
},
"start": {
"type": "boolean",
"title": "Start",
"default": false,
"description": "Start after submission when authorized; false saves a draft for review."
},
"steps": {
"type": "array",
"items": {
"$ref": "#/$defs/Step"
},
"title": "Steps",
"maxItems": 32,
"minItems": 1,
"description": "Production steps in dependency order, with studio arguments and result bindings."
},
"title": {
"type": "string",
"title": "Title",
"maxLength": 160,
"minLength": 1,
"description": "Display title for the plan or movie."
},
"dopamine_hooks": {
"type": "array",
"items": {
"type": "object",
"required": [
"id",
"channel",
"where"
],
"properties": {
"id": {
"enum": [
"novelty",
"surprise_reward",
"uncertain_reward",
"small_wins",
"problem_solved",
"aha_moment",
"social_reward",
"winning",
"music_chills",
"exercise_high",
"awe_nature",
"scent_taste_preview",
"shopping_unboxing",
"cliffhanger",
"info_seeking",
"big_decision_closed",
"teamwork_sync",
"tension_release",
"altruism",
"caffeine_arousal",
"thirst_hunger_relief",
"attraction_anticipation",
"thermal_comfort",
"flow_state",
"humor_punchline",
"order_from_chaos",
"insider_secret",
"cuteness",
"planning_anticipation",
"outrage_justice",
"sugar_rush",
"procrastination_relief"
],
"type": "string"
},
"where": {
"type": "string",
"pattern": "^[^\\r\\n]+$",
"maxLength": 160,
"minLength": 4,
"description": "Where and how it appears in the finished work, in one line. Example: \"0~3초: 오븐 문이 열리며 김이 오르는 빵 클로즈업\"."
},
"channel": {
"enum": [
"visual",
"audio",
"situation"
],
"type": "string",
"description": "How the audience meets it: visual (시각), audio (음향) or situation (상황)."
}
},
"additionalProperties": false
},
"maxItems": 3,
"minItems": 1,
"description": "Creative baseline: the dopamine hook(s) you built into the FINAL work. Declare at least one for any plan that creates image, voice, music, video or movie output, after reading dopamine_hooks_get. The server records each as a \"[도파민 요소] …\" line in goal, shown to the user on the dashboard."
},
"idempotency_key": {
"type": "string",
"title": "Idempotency Key",
"maxLength": 128,
"minLength": 8,
"description": "Unique key for the intended operation. Reuse it on network retries."
},
"acknowledge_execution": {
"type": "boolean",
"title": "Acknowledge Execution",
"default": false,
"description": "Confirm the user explicitly approved starting the production plan automatically."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Submit production plan",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}List plans owned by this account when the user asks; do not use a polling loop.
No parameters.
{
"name": "batches_list",
"title": "List production plans",
"description": "List plans owned by this account when the user asks; do not use a polling loop.",
"inputSchema": {
"type": "object",
"properties": {},
"additionalProperties": false
},
"annotations": {
"title": "List production plans",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Read one plan, steps and results when requested. Prefer handing off its URL after submission.
| Parameter | Type | Description |
|---|---|---|
| batch_id * | string | UUID of the production plan. |
{
"name": "batch_get",
"title": "Get production plan",
"description": "Read one plan, steps and results when requested. Prefer handing off its URL after submission.",
"inputSchema": {
"type": "object",
"required": [
"batch_id"
],
"properties": {
"batch_id": {
"type": "string",
"format": "uuid",
"description": "UUID of the production plan."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Get production plan",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Pause future dispatch, resume, or cancel remaining steps on explicit user request. Already submitted work finishes safely and remains visible.
| Parameter | Type | Description |
|---|---|---|
| action * | string | Pause future dispatch, resume the plan, or cancel the remaining steps. |
| batch_id * | string | UUID of the production plan. |
{
"name": "batch_control",
"title": "Pause, resume or cancel plan",
"description": "Pause future dispatch, resume, or cancel remaining steps on explicit user request. Already submitted work finishes safely and remains visible.",
"inputSchema": {
"type": "object",
"required": [
"batch_id",
"action"
],
"properties": {
"action": {
"enum": [
"pause",
"resume",
"cancel"
],
"type": "string",
"description": "Pause future dispatch, resume the plan, or cancel the remaining steps."
},
"batch_id": {
"type": "string",
"format": "uuid",
"description": "UUID of the production plan."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Pause, resume or cancel plan",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": true
}
}On user request, replace arguments of a definitively rejected failed step that has no accepted request ID. Supply expected_arguments from batch_get to prevent stale edits. Revalidates the entire plan, preserves completed work, and pauses; use batch_control resume after repair. Never use for uncertain or accepted jobs.
| Parameter | Type | Description |
|---|---|---|
| batch_id * | string | UUID of the production plan. |
| step_key * | string | Key of the failed step to repair. |
| arguments * | object | Complete replacement arguments for the failed step. |
| acknowledge * | boolean | Confirm the user authorized repairing this failed step. |
| expected_arguments * | object | Current step arguments from batch_get, used to detect concurrent edits. |
{
"name": "batch_repair",
"title": "Repair failed plan step",
"description": "On user request, replace arguments of a definitively rejected failed step that has no accepted request ID. Supply expected_arguments from batch_get to prevent stale edits. Revalidates the entire plan, preserves completed work, and pauses; use batch_control resume after repair. Never use for uncertain or accepted jobs.",
"inputSchema": {
"type": "object",
"title": "Repair",
"required": [
"batch_id",
"step_key",
"expected_arguments",
"arguments",
"acknowledge"
],
"properties": {
"batch_id": {
"type": "string",
"title": "Batch Id",
"format": "uuid",
"description": "UUID of the production plan."
},
"step_key": {
"type": "string",
"title": "Step Key",
"pattern": "^[a-zA-Z][a-zA-Z0-9_-]{0,39}$",
"description": "Key of the failed step to repair."
},
"arguments": {
"type": "object",
"title": "Arguments",
"description": "Complete replacement arguments for the failed step.",
"additionalProperties": true
},
"acknowledge": {
"type": "boolean",
"const": true,
"title": "Acknowledge",
"description": "Confirm the user authorized repairing this failed step."
},
"expected_arguments": {
"type": "object",
"title": "Expected Arguments",
"description": "Current step arguments from batch_get, used to detect concurrent edits.",
"additionalProperties": true
}
},
"additionalProperties": false
},
"annotations": {
"title": "Repair failed plan step",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}List the account's media library (own recordings and referenced YouTube/TikTok/Instagram Reels/web videos) with status, tags, summary and reuse_policy. Respect reuse_policy: "reusable" footage (own/licensed) may be cut into new work; "reference_only" third-party media is for analysis, structure and style reference only, never copied into an output.
| Parameter | Type | Description |
|---|---|---|
| tag | string | Filter entries by their saved tag. |
| limit | integer | Maximum number of entries or matches to return. |
| query | string | Search text for matching media library entries or moments. |
| origin | string | Filter to a source platform or the user’s own recording. |
| reuse_policy | string | Filter by the library’s recorded footage reuse policy. |
{
"name": "media_library_list",
"title": "List media library",
"description": "List the account's media library (own recordings and referenced YouTube/TikTok/Instagram Reels/web videos) with status, tags, summary and reuse_policy. Respect reuse_policy: \"reusable\" footage (own/licensed) may be cut into new work; \"reference_only\" third-party media is for analysis, structure and style reference only, never copied into an output.",
"inputSchema": {
"type": "object",
"required": [],
"properties": {
"tag": {
"type": "string",
"maxLength": 40,
"description": "Filter entries by their saved tag."
},
"limit": {
"type": "integer",
"maximum": 100,
"minimum": 1,
"description": "Maximum number of entries or matches to return."
},
"query": {
"type": "string",
"maxLength": 200,
"description": "Search text for matching media library entries or moments."
},
"origin": {
"enum": [
"self_recorded",
"youtube",
"tiktok",
"instagram",
"web",
"other"
],
"type": "string",
"description": "Filter to a source platform or the user’s own recording."
},
"reuse_policy": {
"enum": [
"reusable",
"reference_only"
],
"type": "string",
"description": "Filter by the library’s recorded footage reuse policy."
}
},
"additionalProperties": false
},
"annotations": {
"title": "List media library",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Find moments across the library by speech, on-screen text, book-page text or visual similarity (multilingual CLIP). OCR includes book pages; book_page searches only book text and observed printed page labels. Returns item_id, start/end seconds, matched text, page_index (zero-based observation, not printed folio), printed_page and keyframe paths when available. Use for production planning and to pick clip ranges. Respect reuse_policy: "reusable" footage (own/licensed) may be cut into new work; "reference_only" third-party media is for analysis, structure and style reference only, never copied into an output.
| Parameter | Type | Description |
|---|---|---|
| kinds | array | Choose which indexes to search: speech, OCR, visuals or book-page text. |
| limit | integer | Maximum number of entries or matches to return. |
| query * | string | Search text for matching media library entries or moments. |
| reuse_policy | string | Filter by the library’s recorded footage reuse policy. |
{
"name": "media_library_search",
"title": "Search media library",
"description": "Find moments across the library by speech, on-screen text, book-page text or visual similarity (multilingual CLIP). OCR includes book pages; book_page searches only book text and observed printed page labels. Returns item_id, start/end seconds, matched text, page_index (zero-based observation, not printed folio), printed_page and keyframe paths when available. Use for production planning and to pick clip ranges. Respect reuse_policy: \"reusable\" footage (own/licensed) may be cut into new work; \"reference_only\" third-party media is for analysis, structure and style reference only, never copied into an output.",
"inputSchema": {
"type": "object",
"required": [
"query"
],
"properties": {
"kinds": {
"type": "array",
"items": {
"enum": [
"transcript",
"ocr",
"visual",
"book_page"
],
"type": "string"
},
"maxItems": 4,
"description": "Choose which indexes to search: speech, OCR, visuals or book-page text."
},
"limit": {
"type": "integer",
"maximum": 50,
"minimum": 1,
"description": "Maximum number of entries or matches to return."
},
"query": {
"type": "string",
"maxLength": 300,
"minLength": 1,
"description": "Search text for matching media library entries or moments."
},
"reuse_policy": {
"enum": [
"reusable",
"reference_only"
],
"type": "string",
"description": "Filter by the library’s recorded footage reuse policy."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Search media library",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Read one library item's AI-ready manifest (schema hiruai.media-library.v1): provenance and reuse guidance, renditions, catalog (summary, chapters, reusable_segments, voice_reference_candidates, generation_prompts, sensitive flags), scenes with spoken text and OCR. Book-mode items also include book_pages (ordered OCR, observed printed_page, source timestamp/frame, confidence and warnings) and book_analysis coverage limits. Never infer missing pages or treat OCR scores as verified accuracy. include_transcript adds every timed transcript segment. Respect reuse_policy: "reusable" footage (own/licensed) may be cut into new work; "reference_only" third-party media is for analysis, structure and style reference only, never copied into an output.
| Parameter | Type | Description |
|---|---|---|
| item_id * | string | UUID of the media library item. |
| include_transcript | boolean | Include every timed transcript segment in the manifest. |
{
"name": "media_library_get",
"title": "Get media library item",
"description": "Read one library item's AI-ready manifest (schema hiruai.media-library.v1): provenance and reuse guidance, renditions, catalog (summary, chapters, reusable_segments, voice_reference_candidates, generation_prompts, sensitive flags), scenes with spoken text and OCR. Book-mode items also include book_pages (ordered OCR, observed printed_page, source timestamp/frame, confidence and warnings) and book_analysis coverage limits. Never infer missing pages or treat OCR scores as verified accuracy. include_transcript adds every timed transcript segment. Respect reuse_policy: \"reusable\" footage (own/licensed) may be cut into new work; \"reference_only\" third-party media is for analysis, structure and style reference only, never copied into an output.",
"inputSchema": {
"type": "object",
"required": [
"item_id"
],
"properties": {
"item_id": {
"type": "string",
"format": "uuid",
"description": "UUID of the media library item."
},
"include_transcript": {
"type": "boolean",
"description": "Include every timed transcript segment in the manifest."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Get media library item",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Cut a frame-accurate H.264/AAC clip (max 180 s, location metadata removed) from a library item and, by default, save it to your private upload library for up to 30 days. Saving a clip may delete the oldest uploads that unfinished plans no longer need to make room. The returned upload asset_id can be used directly in Video uploads or as a Movie clip (service=upload). Only for reuse_policy=reusable items unless the user confirms they hold the rights.
| Parameter | Type | Description |
|---|---|---|
| item_id * | string | UUID of the media library item. |
| rendition | string | Select the master or proxy as the source, or let the server choose. |
| max_height | integer | Maximum height in pixels for the clipped output. |
| end_seconds * | number | Clip end offset in seconds from the beginning of the source. |
| include_audio | boolean | Include the source audio in the clip when available. |
| save_as_asset | boolean | Save the clip to the private upload library for reuse in subsequent jobs. |
| start_seconds * | number | Clip start offset in seconds from the beginning of the source. |
| idempotency_key * | string | Unique key for the intended operation. Reuse it on network retries. |
| confirm_reference_rights | boolean | Required true to cut a reference_only item; the user must state they hold permission. |
{
"name": "media_library_clip",
"title": "Cut clip from library item",
"description": "Cut a frame-accurate H.264/AAC clip (max 180 s, location metadata removed) from a library item and, by default, save it to your private upload library for up to 30 days. Saving a clip may delete the oldest uploads that unfinished plans no longer need to make room. The returned upload asset_id can be used directly in Video uploads or as a Movie clip (service=upload). Only for reuse_policy=reusable items unless the user confirms they hold the rights.",
"inputSchema": {
"type": "object",
"required": [
"item_id",
"start_seconds",
"end_seconds",
"idempotency_key"
],
"properties": {
"item_id": {
"type": "string",
"format": "uuid",
"description": "UUID of the media library item."
},
"rendition": {
"enum": [
"auto",
"master",
"proxy"
],
"type": "string",
"description": "Select the master or proxy as the source, or let the server choose."
},
"max_height": {
"enum": [
480,
720,
1080
],
"type": "integer",
"description": "Maximum height in pixels for the clipped output."
},
"end_seconds": {
"type": "number",
"description": "Clip end offset in seconds from the beginning of the source.",
"exclusiveMinimum": 0
},
"include_audio": {
"type": "boolean",
"description": "Include the source audio in the clip when available."
},
"save_as_asset": {
"type": "boolean",
"description": "Save the clip to the private upload library for reuse in subsequent jobs."
},
"start_seconds": {
"type": "number",
"minimum": 0,
"description": "Clip start offset in seconds from the beginning of the source."
},
"idempotency_key": {
"type": "string",
"maxLength": 128,
"minLength": 8,
"description": "Unique key for the intended operation. Reuse it on network retries."
},
"confirm_reference_rights": {
"type": "boolean",
"description": "Required true to cut a reference_only item; the user must state they hold permission."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Cut clip from library item",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": true
}
}Read the creative baseline catalog: everyday situations in which people feel reward and anticipation (dopamine hooks), each with visual, audio and situation cues and guardrails, plus the planning principle and this week's practice theme. Call before planning any creative work, then build at least one hook into the final output and declare it in batch_validate/batch_submit. Filter with group or ids. No generation, quota or GPU job.
| Parameter | Type | Description |
|---|---|---|
| ids | array | Filter hooks to the specified catalog IDs. |
| group | string | Filter hooks to one catalog group. |
{
"name": "dopamine_hooks_get",
"title": "Read dopamine hooks",
"description": "Read the creative baseline catalog: everyday situations in which people feel reward and anticipation (dopamine hooks), each with visual, audio and situation cues and guardrails, plus the planning principle and this week's practice theme. Call before planning any creative work, then build at least one hook into the final output and declare it in batch_validate/batch_submit. Filter with group or ids. No generation, quota or GPU job.",
"inputSchema": {
"type": "object",
"properties": {
"ids": {
"type": "array",
"items": {
"enum": [
"novelty",
"surprise_reward",
"uncertain_reward",
"small_wins",
"problem_solved",
"aha_moment",
"social_reward",
"winning",
"music_chills",
"exercise_high",
"awe_nature",
"scent_taste_preview",
"shopping_unboxing",
"cliffhanger",
"info_seeking",
"big_decision_closed",
"teamwork_sync",
"tension_release",
"altruism",
"caffeine_arousal",
"thirst_hunger_relief",
"attraction_anticipation",
"thermal_comfort",
"flow_state",
"humor_punchline",
"order_from_chaos",
"insider_secret",
"cuteness",
"planning_anticipation",
"outrage_justice",
"sugar_rush",
"procrastination_relief"
],
"type": "string"
},
"maxItems": 32,
"description": "Filter hooks to the specified catalog IDs."
},
"group": {
"enum": [
"everyday",
"biological",
"flow",
"social",
"caution"
],
"type": "string",
"description": "Filter hooks to one catalog group."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Read dopamine hooks",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Read the deployed Video reference paths and GPU validation status. Distinguishes endpoint keyframes from full-video depth/canny/pose IC-LoRA. No generation.
No parameters.
{
"name": "video_reference_capabilities",
"title": "Read video reference support",
"description": "Read the deployed Video reference paths and GPU validation status. Distinguishes endpoint keyframes from full-video depth/canny/pose IC-LoRA. No generation.",
"inputSchema": {
"type": "object",
"properties": {},
"additionalProperties": false
},
"annotations": {
"title": "Read video reference support",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}You can list tools, resources, resource templates and prompts before signing in. Calling a tool, reading a resource or getting a prompt requires OAuth sign-in. Creative baseline: no separate skill installation is required. Start with workflow_get(topic=start), then the matching recipe and capabilities_get(service=...). For quality-first image generation or editing, prefer the connected agent’s own image tool when it supports the requested operation, for example ChatGPT image generation and image editing. Check what this client actually exposes, including in Claude. Honor the user’s chosen provider. For anime-style character illustration, HiruAI Image offers WAI Illustrious SDXL v17 when available in image_models_get. Read workflow_get(topic=image) before preparing stills. When one Video shot continues another, use the previous clip’s actual final frame as the next opening reference. Read workflow_get(recipe_id=video_continuity) for the supported Movie-to-Video bindings. Read live schemas and obtain actual source/voice consent before writes. Validate and submit the whole plan once within scope; return its dashboard URL. The server continues after the chat ends; do not poll. Recipes are available through tools; resources and prompts are optional. Creative baseline, the default template for every creative request: the connected agent plans and this server produces. Before writing prompts, scripts or edits, read dopamine_hooks_get (or resource hiruai://creative-cloud/dopamine-hooks) and build AT LEAST ONE dopamine hook into the final work so the audience sees it, hears it or feels it as a situation; describing it only in the plan does not count. Choose what fits the user's subject and audience, set up the expectation early and let the payoff land a little later, and prefer one or two clear hooks over many faint ones. Tell the user in one line which hook you used and where, and declare it as dopamine_hooks=[{id,channel,where}] in batch_validate and batch_submit. Follow each hook's guardrail and use only facts the user gave: never invent discounts, prizes, reviews, numbers or awards. If the user declines hooks, proceed without one and say so. The server does not choose or generate the hook. Movie: read movie_editing.duration_planning in capabilities_get. For exact shorts use fixed + padding_policy=shrink; default expand permits longer output. batch_validate is static validation; measured timing is checked after Voice completes. Never infer speech timestamps. Input contract: read each studio schema, including maxLength/minLength, enum, numeric bounds and conditional input requirements. Music time_signature is "4" for 4/4, never "4/4" (allowed "2","3","4","6"). Validate the complete plan with batch_validate; fix every reported issue before batch_submit. Submission and dispatch independently enforce the same contract. Validation errors include the step/field, rule, limit or allowed values and a correction example. For failed unaccepted steps, batch_repair preserves earlier completed results. Local files: call asset_upload first (or use the private web file library), then reference returned IDs in uploads=[{field,asset_id}] with upload mode for Video/Edit/Upscale/Voice. Movie accepts service=upload assets directly. Never publish files just to make a batch work. Read capabilities_get and assets_list. Submit a complete plan with batch_submit, then give its URL to the user and STOP; no completion waiting, repeated request_get, or scheduled LLM polling. The server persists and advances the plan without an active chat or OAuth session. Order steps topologically; bindings inject a completed producer request id or Movie asset into a nested arguments path (e.g. image_request_id, generated_request_id.0, clips.0.asset). Default start=false requires web review; start=true needs explicit execution authorization. Rights/privacy/safety acknowledgements must reflect user consent. Standard accounts allow one active request per studio through result ema
Names, descriptions and schemas were written by the server and are shown as data.