このMCPでできること
Generates and edits branded images, video, audio, speech, and 3D-related creative assets, with media processing and reusable themes.
ツール
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['video_url'], 'properties': {'language': {'type': 'string', 'description': "ISO-639 language code for transcription (e.g., 'en', 'es', 'fr', 'de', 'ja', 'zh', 'ko'). Send it only when the user specifies the language or the spoken language is reliably known; otherwise omit it. Use 'zh' for Chinese; provider-specific normalization is handled internally. Default: en."}, 'position': {'enum': ['top', 'center', 'bottom'], 'type': 'string', 'description': 'Vertical subtitle position. Defaults to bottom; explicit choices are preserved on every aspect ratio.'}, 'y_offset': {'type': 'integer', 'maximum': 200, 'minimum': -200, 'description': 'Bounded vertical adjustment in pixels. With bottom placement, this is a positive inward margin from the bottom edge (20-200); larger values place captions higher. Center placement accepts -200 to 200. Defaults to a frame-aware safe margin.'}, 'font_name': {'type': 'string', 'description': 'Optional exact Google Fonts family name. Do not set font_name unless the user explicitly requests a specific font; never invent or infer one. When omitted, Creative Claw chooses the matching Noto Sans family for the language. Explicit fonts are checked for language-script coverage before credits are charged and rejected if incompatible.'}, 'font_size': {'type': 'integer', 'maximum': 150, 'minimum': 20, 'description': 'Preferred subtitle size in pixels (20-150). It is kept within a frame-aware safe range; omit it for a large social-caption default.'}, 'video_url': {'type': 'string', 'description': 'URL of the video to add subtitles to'}, 'font_color': {'enum': ['white', 'black', 'red', 'green', 'blue', 'yellow', 'orange', 'purple', 'pink', 'brown', 'gray', 'cyan', 'magenta'], 'type': 'string', 'description': 'Text color for non-active words. Default: white. Omit unless the user requests a color or you have reliable visual context and are confident another supported color will read better over this video.'}, 'font_weight': {'enum': ['normal', 'bold', 'black'], 'type': 'string', 'description': 'Font weight. Default: bold'}, 'stroke_color': {'enum': ['white', 'black', 'red', 'green', 'blue', 'yellow', 'orange', 'purple', 'pink', 'brown', 'gray', 'cyan', 'magenta'], 'type': 'string', 'description': 'Text stroke/outline color. Default: black'}, 'stroke_width': {'type': 'integer', 'maximum': 10, 'minimum': 0, 'description': 'Text stroke/outline width in pixels, 0 for no stroke. Default: 3'}, 'highlight_color': {'enum': ['white', 'black', 'red', 'green', 'blue', 'yellow', 'orange', 'purple', 'pink', 'brown', 'gray', 'cyan', 'magenta'], 'type': 'string', 'description': 'Color for the currently spoken word only, creating the karaoke cue. Default: purple while non-active words remain white. Omit unless the user requests a color or reliable visual context makes a different supported highlight clearly better.'}, 'background_color': {'enum': ['white', 'black', 'red', 'green', 'blue', 'yellow', 'orange', 'purple', 'pink', 'brown', 'gray', 'cyan', 'magenta', 'none', 'transparent'], 'type': 'string', 'description': 'Background color behind subtitle text. Default: none'}, 'enable_animation': {'type': 'boolean', 'description': 'Enable bounce-style entrance animation. Default: false.'}, 'background_opacity': {'type': 'number', 'maximum': 1, 'minimum': 0, 'description': 'Background opacity 0.0 to 1.0'}, 'words_per_subtitle': {'type': 'integer', 'maximum': 12, 'description': 'Preferred max words per subtitle segment. Portrait video uses 3 words per segment, square is capped at 1, and landscape is capped at 3.', 'exclusiveMinimum': 0}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['id'], 'properties': {'id': {'type': 'string', 'format': 'uuid', 'pattern': '^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$', 'description': 'Existing Film project id returned by create_film_project or list_film_projects; never use a placeholder UUID'}, 'mode': {'enum': ['connect', 'cut_end'], 'type': 'string', 'default': 'connect', 'description': 'connect preserves every clip and warns about target overage; cut_end trims the fully assembled output at the target duration.'}, 'video_fit': {'enum': ['auto', 'crop', 'pad', 'strict'], 'type': 'string', 'default': 'crop', 'description': "Fit clips to the first clip's canvas. crop (default) and auto center-crop mismatches; pad preserves the full frame with black bars; strict rejects mismatches. Original shot URLs are preserved. Completed crops are reused on later assemblies when available. Pricing: 2 base credits plus 1 per distinct uncached normalization; one transient crop retry adds no charge."}, 'with_narration': {'type': 'boolean', 'default': True, 'description': "Lay the project's audioUrl narration track over the stitched video, if present."}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['job_id'], 'properties': {'job_id': {'type': 'string', 'description': 'Job ID returned by an async generation or media utility tool'}, 'ui_refresh': {'type': 'boolean', 'description': 'Set to true only for automatic status checks from the media viewer UI.'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['consent'], 'properties': {'consent': {'type': 'boolean', 'description': "Must be true after the user explicitly agrees to voice cloning and processing by supported voice providers selected through Creative Claw. Wording: I own this recording or have permission to use it, and this is my voice or I have the speaker's explicit permission to clone it and generate speech in this workspace. I authorize Creative Claw and the supported voice providers I select through Creative Claw to process it for this purpose, and accept the voice-cloning terms and privacy notice. Terms: https://creativeclaw.co/terms/#voice-cloning Privacy: https://creativeclaw.co/privacy/#voice-cloning"}, 'language': {'type': 'string', 'default': 'en', 'maxLength': 16, 'minLength': 2, 'description': 'Source recording language code, such as en, es, or he. Used immediately for Cartesia and retained for a future lazy Cartesia copy when ElevenLabs is cloned first.'}, 'provider': {'enum': ['elevenlabs', 'cartesia'], 'type': 'string', 'default': 'elevenlabs', 'description': 'Voice provider. Defaults to elevenlabs for compatibility. Choose cartesia explicitly to create a Cartesia clone; providers are never switched silently.'}, 'audio_url': {'type': 'string', 'format': 'uri', 'description': "A publicly downloadable audio link. Creative Claw copies the original bytes into this workspace's private voice storage before cloning. Private-network targets and non-audio files are rejected. Prefer audio_asset_id when the recording is already privately imported."}, 'character_id': {'type': 'string', 'format': 'uuid', 'pattern': '^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$', 'description': 'Existing Character to receive the cloned voice. Omit to create a reusable voice-only Character automatically.'}, 'audio_asset_id': {'type': 'string', 'format': 'uuid', 'pattern': '^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$', 'description': "Private voice sample asset ID. For a ChatGPT attachment call import_chatgpt_media({ media_file: <attached file>, purpose: 'voice_clone' }); otherwise use import_media({ purpose: 'voice_clone' }) or get_upload_url({ type: 'audio', purpose: 'voice_clone', content_type }) followed by confirm_upload. Public media assets cannot be used."}, 'character_name': {'type': 'string', 'maxLength': 120, 'minLength': 1, 'description': 'Name for an automatically created voice-only Character. Used only when character_id is omitted; defaults to "My Voice".'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['prompt', 'models'], 'properties': {'seed': {'type': 'number', 'description': 'Seed for reproducibility (shared)'}, 'width': {'type': 'number', 'description': 'Image width in pixels (shared)'}, 'height': {'type': 'number', 'description': 'Image height in pixels (shared)'}, 'models': {'type': 'array', 'items': {'type': 'string'}, 'maxItems': 4, 'minItems': 2, 'description': 'Array of 2-4 model IDs to compare (e.g. ["image/gpt-image-2.5-flare", "image/gpt-image-2.5-sunburst", "image/nano-banana-2"])'}, 'prompt': {'type': 'string', 'description': 'Shared text prompt for all models'}, 'output_format': {'enum': ['jpeg', 'png'], 'type': 'string', 'default': 'jpeg', 'description': 'Output format (shared)'}, 'negative_prompt': {'type': 'string', 'description': 'Elements to exclude (shared)'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['asset_id'], 'properties': {'asset_id': {'type': 'string', 'format': 'uuid', 'pattern': '^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$', 'description': 'The asset ID returned by get_upload_url'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['name'], 'properties': {'name': {'type': 'string', 'description': 'Short project title'}, 'brief': {'type': 'string', 'description': "The user's creative brief / what the film is about"}, 'theme_id': {'type': 'string', 'format': 'uuid', 'pattern': '^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$', 'description': 'Brand theme id'}, 'character_ids': {'type': 'array', 'items': {'type': 'string', 'format': 'uuid', 'pattern': '^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$'}, 'description': 'Character ids for the cast (from list_characters)'}, 'target_duration_s': {'type': 'integer', 'maximum': 3600, 'minimum': 4, 'description': 'Target total film length in seconds, up to 60 minutes; shot duration is model-dependent'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['name'], 'properties': {'html': {'type': 'string', 'description': "HTML markup with {{token}} placeholders for parameters. Required when kind='html'. Conditional blocks supported via Mustache syntax: {{#booleanParam}}...{{/booleanParam}} and {{^booleanParam}}...{{/booleanParam}}."}, 'kind': {'enum': ['html', 'generative'], 'type': 'string', 'default': 'html', 'description': "Template kind. 'html' renders HTML/CSS via headless Chromium â\x86\x92 PNG. 'generative' fills a prompt template and calls a generative model."}, 'name': {'type': 'string', 'minLength': 1, 'description': 'Human-readable template name'}, 'sizes': {'type': 'array', 'items': {'type': 'object', 'required': ['name', 'width', 'height'], 'properties': {'name': {'type': 'string', 'pattern': '^[a-zA-Z0-9_-]+$', 'description': "Identifier for this size â\x80\x94 used in render_template's size_names. e.g. 'story', 'square', '16x9'."}, 'label': {'type': 'string', 'description': "Human-readable label, e.g. 'Instagram Story'."}, 'width': {'type': 'integer', 'maximum': 9007199254740991, 'exclusiveMinimum': 0}, 'height': {'type': 'integer', 'maximum': 9007199254740991, 'exclusiveMinimum': 0}, 'bodyClass': {'type': 'string', 'pattern': '^[\\w\\s-]+$', 'description': "CSS class appended to <body> when this size is rendered. Lets the template's <style> use selectors like `body.story .hero { ... }` to vary layout per size."}, 'isDefault': {'type': 'boolean', 'description': 'Mark as the default size used when render_template is called without size_names.'}}}, 'description': "html kind only. Named output sizes the template can be rendered at (e.g. 9:16 story + 1:1 square). The same HTML body is rendered at every size; per-size CSS overrides go through each size's `bodyClass` (selectors like `body.story .subtitle { display: none }`). Exactly one size should set isDefault=true. If omitted, falls back to a single 'default' size derived from width/height."}, 'width': {'type': 'integer', 'maximum': 9007199254740991, 'description': "Default output width in pixels. Required when kind='html' and `sizes` is not provided.", 'exclusiveMinimum': 0}, 'height': {'type': 'integer', 'maximum': 9007199254740991, 'description': "Default output height in pixels. Required when kind='html' and `sizes` is not provided.", 'exclusiveMinimum': 0}, 'prompt': {'type': 'string', 'description': "Generative kind only. Prompt body with {{token}} placeholders. Required when kind='generative'."}, 'media_type': {'enum': ['image', 'video'], 'type': 'string', 'description': "Generative kind only. Output media type. Required when kind='generative'."}, 'parameters': {'type': 'array', 'items': {'type': 'object', 'required': ['name', 'type'], 'properties': {'name': {'type': 'string', 'pattern': '^[a-zA-Z_][a-zA-Z0-9_]*$'}, 'type': {'enum': ['text', 'image_url', 'color', 'number', 'boolean'], 'type': 'string'}, 'default': {'type': 'string'}, 'required': {'type': 'boolean'}, 'description': {'type': 'string'}}}, 'default': [], 'description': "Schema for {{token}} variable inputs. Each parameter has a name that matches a token in the html or prompt body. Types: 'text', 'image_url', 'color', 'number', 'boolean'. Boolean params can also drive Mustache-style conditional blocks: {{#name}}...{{/name}} renders the block when truthy, {{^name}}...{{/name}} when falsy."}, 'description': {'type': 'string', 'description': 'Short description of what this template is for'}, 'model_params': {'type': 'object', 'description': 'Generative kind only. Default model params (size, aspect_ratio, etc.) merged into the model call.', 'propertyNames': {'type': 'string'}, 'additionalProperties': {}}, 'recommended_model': {'type': 'string', 'description': "Generative kind only. Registry model id (e.g. 'image/nano-banana-pro'). Required when kind='generative'."}, 'reference_asset_ids': {'type': 'array', 'items': {'type': 'string', 'format': 'uuid', 'pattern': '^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$'}, 'description': 'Generative kind only. Asset IDs of reference images used as visual examples / image_urls.'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['video_url', 'segments'], 'properties': {'fps': {'enum': ['24', '25', '30', '50', '60', 'source'], 'type': 'string', 'default': 'source'}, 'name': {'type': 'string', 'maxLength': 200}, 'tags': {'type': 'array', 'items': {'type': 'string', 'maxLength': 100}, 'maxItems': 30}, 'width': {'type': 'integer', 'default': 1080, 'maximum': 1920, 'minimum': 128, 'description': 'Output width in pixels. Any even value from 128-1920; defaults to 1080 for 9:16.'}, 'height': {'type': 'integer', 'default': 1920, 'maximum': 1920, 'minimum': 128, 'description': 'Output height in pixels. Any even value from 128-1920; defaults to 1920 for 9:16.'}, 'captions': {'type': 'object', 'required': ['words'], 'properties': {'font': {'enum': ['Noto Sans', 'Noto Sans Hebrew', 'Noto Sans Arabic', 'Noto Sans CJK SC', 'Noto Sans CJK JP', 'Noto Sans CJK KR'], 'type': 'string', 'default': 'Noto Sans'}, 'style': {'enum': ['plain', 'karaoke'], 'type': 'string', 'default': 'plain'}, 'words': {'type': 'array', 'items': {'type': 'object', 'required': ['start', 'end', 'text'], 'properties': {'end': {'type': 'number', 'maximum': 14400, 'minimum': 0}, 'text': {'type': 'string', 'maxLength': 100, 'minLength': 1}, 'start': {'type': 'number', 'maximum': 14400, 'minimum': 0}}, 'additionalProperties': False}, 'maxItems': 10000, 'minItems': 1}, 'words_per_caption': {'type': 'integer', 'default': 3, 'maximum': 100, 'minimum': 1, 'description': 'Target words per caption. Values above 7 are normalized to 7.'}}, 'description': 'Optional burned captions. Omit for no new captions. Uses source-timed words, not output timestamps; words that straddle a cut are omitted while the requested video cut is preserved. Original speech is preserved.', 'additionalProperties': False}, 'segments': {'type': 'array', 'items': {'type': 'object', 'required': ['start', 'end'], 'properties': {'end': {'type': 'number', 'maximum': 14400, 'minimum': 0}, 'start': {'type': 'number', 'maximum': 14400, 'minimum': 0}, 'framing': {'type': 'object', 'required': ['mode'], 'properties': {'x': {'type': 'number', 'maximum': 14400, 'minimum': 0, 'description': 'Static crop x coordinate in source pixels; crop mode only.'}, 'y': {'type': 'number', 'maximum': 14400, 'minimum': 0, 'description': 'Static crop y coordinate in source pixels; crop mode only.'}, 'mode': {'enum': ['pad', 'center_crop', 'crop'], 'type': 'string', 'description': 'pad and center_crop take no geometry. Only crop accepts width, height, static x/y, or keyframes.'}, 'width': {'type': 'integer', 'maximum': 9007199254740991, 'description': 'Crop width in source pixels; crop mode only.', 'exclusiveMinimum': 0}, 'height': {'type': 'integer', 'maximum': 9007199254740991, 'description': 'Crop height in source pixels; crop mode only.', 'exclusiveMinimum': 0}, 'keyframes': {'type': 'array', 'items': {'type': 'object', 'required': ['time', 'x', 'y'], 'properties': {'x': {'type': 'number', 'maximum': 14400, 'minimum': 0}, 'y': {'type': 'number', 'maximum': 14400, 'minimum': 0}, 'time': {'type': 'number', 'maximum': 14400, 'minimum': 0}}, 'additionalProperties': False}, 'maxItems': 20, 'minItems': 1, 'description': 'Moving crop path; crop mode only. Do not combine with static x/y.'}}, 'description': 'Framing mode. Use exactly {"mode":"pad"} or {"mode":"center_crop"} for those modes. Crop geometry must match the output aspect ratio.', 'additionalProperties': False}}, 'additionalProperties': False}, 'maxItems': 40, 'minItems': 1, 'description': 'Ordered half-open source ranges in seconds. Non-overlapping, at least 150ms each. Framing uses display-oriented source pixels; keyframe times are segment-relative. No automatic face tracking.'}, 'video_url': {'type': 'string', 'format': 'uri', 'description': 'Existing video asset URL owned by this workspace; import the file first.'}, 'audio_fade_ms': {'type': 'number', 'default': 30, 'maximum': 1000, 'minimum': 0, 'description': 'Boundary fade in milliseconds for discontinuous audio cuts. Defaults to 30. Values above 100 are accepted for client compatibility and normalized to 100.'}, 'normalize_audio': {'type': 'boolean', 'default': False}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['id'], 'properties': {'id': {'type': 'string', 'format': 'uuid', 'pattern': '^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$', 'description': 'The asset ID to delete'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['id'], 'properties': {'id': {'type': 'string', 'format': 'uuid', 'pattern': '^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$', 'description': 'ID of the character to delete'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['name'], 'properties': {'name': {'type': 'string', 'description': 'Name of the theme to delete'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['operation'], 'properties': {'model': {'type': 'string', 'description': 'Model ID to estimate. Omit for html_video renders and video_upscale utilities.'}, 'params': {'type': 'object', 'default': {}, 'description': 'The exact generation parameters being considered, including duration, resolution, image count, text, references, and model-specific extras.', 'propertyNames': {'type': 'string'}, 'additionalProperties': {}}, 'operation': {'enum': ['image', 'video', 'speech', 'audio', 'html_video', 'video_upscale'], 'type': 'string', 'description': 'Generation type to estimate.'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['video_url', 'mode'], 'properties': {'mode': {'enum': ['single', 'batch'], 'type': 'string', 'description': "Extraction mode. 'single' extracts one frame (first/middle/last) from videos up to 500 MiB. 'batch' extracts every Nth frame from MP4/MOV videos smaller than 100 MB."}, 'quality': {'type': 'integer', 'maximum': 100, 'minimum': 1, 'description': 'Quality for jpg/webp output (for batch mode, 1-100). Default: 95'}, 'position': {'enum': ['first', 'middle', 'last'], 'type': 'string', 'description': 'Which frame to extract (for single mode). Default: middle'}, 'video_url': {'type': 'string', 'description': 'URL of the video to extract frames from'}, 'max_frames': {'type': 'integer', 'maximum': 500, 'minimum': 1, 'description': 'Batch mode only: stop after this many extracted frames (1-500). Every returned frame is uploaded and included in the response, so use a small value when possible. Default: 100'}, 'output_format': {'enum': ['png', 'jpg', 'webp'], 'type': 'string', 'description': 'Image format for extracted frames (for batch mode). Default: png'}, 'frame_interval': {'type': 'integer', 'maximum': 300, 'minimum': 1, 'description': 'Batch mode only: extract every Nth video frame, where N is 1-300 (not a timestamp or number of seconds). For example, 30 extracts about once per second from a 30fps video. Default: 12'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['prompt'], 'properties': {'seed': {'type': 'number', 'description': 'Seed for reproducible results'}, 'size': {'enum': ['1:1', '4:5', '5:4', '9:16', '16:9'], 'type': 'string', 'description': 'Output aspect ratio. Strongly recommended over width/height â\x80\x94 these are the sizes our supported models render reliably. 1:1 = square (1080x1080), 4:5 = Instagram portrait (1080x1350), 5:4 = landscape (1350x1080), 9:16 = story/reel (1080x1920), 16:9 = wide (1920x1080).'}, 'model': {'type': 'string', 'default': 'image/nano-banana-2', 'description': 'Model ID to use. Default and cost-efficient top pick: image/nano-banana-2. OpenAI routes: image/gpt-image-2.5-flare for speed and image/gpt-image-2.5-sunburst for precision. Other specialists include image/nano-banana-pro and image/seedream-5-pro.'}, 'width': {'type': 'number', 'description': 'Image width in pixels. Prefer the `size` field â\x80\x94 width/height is kept for backwards compatibility.'}, 'extras': {'type': 'object', 'description': 'Additional model-specific parameters. Use get_model_params to discover available extras for a model.', 'propertyNames': {'type': 'string'}, 'additionalProperties': {}}, 'height': {'type': 'number', 'description': 'Image height in pixels. Prefer the `size` field â\x80\x94 width/height is kept for backwards compatibility.'}, 'prompt': {'type': 'string', 'description': 'Text description of the image to generate, or edit instructions when image_url is provided'}, 'strength': {'type': 'number', 'maximum': 1, 'minimum': 0, 'description': 'Transformation strength when editing (0 = no change, 1 = full transformation). Default: 0.75'}, 'image_url': {'type': 'string', 'description': 'Source image URL for editing. Must be public/directly fetchable or a Creative Claw URL. In ChatGPT, call import_chatgpt_media for a file already pasted, attached, or generated in the conversation; call import_media only to open the upload picker.'}, 'num_images': {'type': 'number', 'default': 1, 'maximum': 4, 'minimum': 1, 'description': 'Number of images to generate (1-4)'}, 'character_id': {'type': 'string', 'format': 'uuid', 'pattern': '^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$', 'description': 'Select a saved Character (persona). Its description is woven into the prompt and its reference image is used as a visual anchor (source/reference) so the character stays consistent. Use list_characters to find ids.'}, 'output_format': {'enum': ['jpeg', 'png'], 'type': 'string', 'default': 'jpeg', 'description': 'Output image format'}, 'guidance_scale': {'type': 'number', 'description': 'How closely to follow the prompt (CFG scale)'}, 'agentic_prompting': {'type': 'boolean', 'description': "Optional Creative Claw prompt rewrite. Omit it (default) or set false to preserve the user's wording; set true only when the user wants a model-specific rewrite. Adds ~1-2s latency. Selected Character identity context is appended deterministically either way."}, 'remove_background': {'type': 'boolean', 'default': False, 'description': 'Remove background from generated image(s). Returns a transparent PNG with the background removed.'}, 'num_inference_steps': {'type': 'number', 'description': 'Number of denoising steps (higher = better quality, slower)'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['prompt'], 'properties': {'model': {'type': 'string', 'const': 'music/elevenlabs-music-v2.5', 'default': 'music/elevenlabs-music-v2.5', 'description': 'Music model. Currently music/elevenlabs-music-v2.5.'}, 'prompt': {'type': 'string', 'maxLength': 4100, 'minLength': 1, 'description': 'Describe the genre, tempo, instruments, mood, structure, mix, ending, and vocal role. Maximum 4,100 characters.'}, 'output_format': {'enum': ['mp3_44100_128', 'mp3_44100_192', 'mp3_48000_192', 'pcm_44100', 'pcm_48000'], 'type': 'string', 'default': 'mp3_48000_192', 'description': 'Audio output format. Supported: mp3_44100_128, mp3_44100_192, mp3_48000_192, pcm_44100, pcm_48000. Default: mp3_48000_192. Unsupported values automatically use the default.'}, 'music_length_ms': {'type': 'integer', 'maximum': 600000, 'minimum': 3000, 'description': 'Track length in milliseconds, from 3,000 to 600,000. Default: 30,000.'}, 'force_instrumental': {'type': 'boolean', 'description': 'Prevent vocals. Default: true. Set false only when vocals are wanted.'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['prompt'], 'properties': {'loop': {'type': 'boolean', 'description': 'Generate a seamless ambience or texture loop.'}, 'model': {'type': 'string', 'const': 'sfx/elevenlabs-sound-v2', 'default': 'sfx/elevenlabs-sound-v2', 'description': 'Sound-effect model. Currently sfx/elevenlabs-sound-v2.'}, 'prompt': {'type': 'string', 'maxLength': 450, 'minLength': 1, 'description': 'Describe the audible source, action, timing, space, texture, intensity, and ending. Maximum 450 characters.'}, 'output_format': {'enum': ['mp3_44100_128', 'mp3_44100_192', 'pcm_44100', 'pcm_48000'], 'type': 'string', 'default': 'mp3_44100_128', 'description': 'Audio output format. Supported: mp3_44100_128, mp3_44100_192, pcm_44100, pcm_48000. Default: mp3_44100_128. Unsupported values automatically use the default.'}, 'duration_seconds': {'type': 'number', 'maximum': 30, 'minimum': 0.5, 'description': 'Duration from 0.5 to 30 seconds. Omit to let ElevenLabs choose.'}, 'prompt_influence': {'type': 'number', 'maximum': 1, 'minimum': 0, 'description': 'Prompt adherence from 0 to 1. Default: 0.3.'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['text'], 'properties': {'text': {'type': 'string', 'description': 'Text to convert to speech. Limits vary by model; xAI TTS accepts up to 15,000 characters.'}, 'model': {'type': 'string', 'default': 'speech/elevenlabs-v3', 'description': 'Model ID for text-to-speech. Use speech/elevenlabs-v3 for stock ElevenLabs voices and general speech. Choose speech/elevenlabs-v2 only for an existing cloned Character voice when a steady read is preferred, or when the user explicitly requests v2. Use speech/cartesia-sonic for fast natural stock or Character speech. Omitting model uses v3.'}, 'speed': {'type': 'number', 'maximum': 2, 'minimum': 0.5, 'description': 'Speech rate (0.5-2.0, default 1.0)'}, 'extras': {'type': 'object', 'description': 'Additional model-specific parameters. Use get_model_params to discover available extras for a model.', 'propertyNames': {'type': 'string'}, 'additionalProperties': {}}, 'format': {'type': 'string', 'description': 'Output audio format: "mp3" (default), "pcm", "flac"'}, 'emotion': {'enum': ['happy', 'sad', 'angry', 'fearful', 'disgusted', 'surprised', 'excited', 'content', 'scared', 'calm', 'fluent', 'whisper', 'neutral'], 'type': 'string', 'description': 'Model-specific global emotion. For Cartesia use neutral, angry, excited, content, sad, or scared. ElevenLabs v3 uses inline audio tags instead. Call get_model_params for the selected model.'}, 'voice_id': {'type': 'string', 'description': 'Stock voice ID. This selects an existing voice; it does not design a voice from prose. generate_speech has no voice_description parameter. For ElevenLabs or Cartesia, call get_model_params and select a curated recommendation matching the requested gender, accent, language, and tone, or pass an exact public provider Voice Library ID. Omitting voice_id for ElevenLabs uses Hale, a smooth, confident American male. For a private cloned voice, pass character_id instead. Never combine an unrelated voice_id and character_id.'}, 'audio_url': {'type': 'string', 'format': 'uri', 'description': 'Reference audio for models that explicitly support it, such as Chatterbox English or Chatterbox Multilingual. For Chatterbox Multilingual, also pass extras.custom_audio_language. For recommended ElevenLabs voice cloning, pass the sample to clone_voice first, then use character_id here. This field does not create an ElevenLabs clone.'}, 'force_new': {'type': 'boolean', 'description': 'Bypass the 10-minute same-workspace duplicate guard and submit a new paid speech generation. Set true only when the user explicitly wants another variation from otherwise identical parameters.'}, 'sample_rate': {'type': 'string', 'description': 'Sample rate: "8000", "16000", "22050", "24000", "32000" (default), "44100"'}, 'character_id': {'type': 'string', 'format': 'uuid', 'pattern': '^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$', 'description': "Character ID with a provider-hosted voice attached through clone_voice. The speech model must match the Character's voice provider. Use list_characters to find it; if it has no voice, call clone_voice first."}, 'language_boost': {'type': 'string', 'description': 'Model-specific language hint, such as "Spanish" for MiniMax. Use only when get_model_params exposes it. ElevenLabs v3 uses extras.language_code (for example, "es"); Multilingual v2 detects language from text and does not support language_code. Chatterbox does not expose language_boost.'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['prompt'], 'properties': {'model': {'type': 'string', 'default': 'video/gemini-omni-flash', 'description': 'Model ID for video generation. Default and top pick: video/gemini-omni-flash. Recommended specialist routes: video/seedance-2.5, video/minimax-h3-max, video/minimax-h3-max-turbo, and Seedance Mini as returned by list_models.'}, 'extras': {'type': 'object', 'description': 'Model-specific options returned under extras by get_model_params. Media fields belong at top level. Legacy media fields here are normalized without reordering; conflicting duplicates or unsupported combinations return corrective errors before charging.', 'propertyNames': {'type': 'string'}, 'additionalProperties': {}}, 'prompt': {'type': 'string', 'description': 'Meaningful video-generation instructions, not a control command or placeholder. Describe at least the visible subject and action, and add camera movement, setting or style, timing, and audio when relevant.'}, 'duration': {'anyOf': [{'type': 'string'}, {'type': 'integer', 'maximum': 9007199254740991, 'minimum': -9007199254740991}], 'description': 'Video duration â\x80\x94 model-dependent. For Seedance 2.5 edits, omit this or use "auto" to preserve the source timeline; Creative Claw sends Pika the required literal "auto" value. Seedance 2.5 reference/image-to-video and extensions use whole seconds from 4 through 30. MiniMax H3 and H3 Max use whole seconds from 5 through 15. Veo: "4s", "6s", "8s". Wan: "5", "10", "15". Use get_model_params for exact current limits.'}, 'force_new': {'type': 'boolean', 'description': 'Bypass the 10-minute same-workspace duplicate guard and submit a new paid generation. Set true only when the user explicitly wants another variation from otherwise identical parameters.'}, 'image_url': {'type': 'string', 'description': 'Literal first-frame image URL only. This selects image-to-video and makes the image frame zero. If an image should instead guide identity, style, character, product, or composition, put it in image_urls, even when there is only one reference image. Do not combine with reference arrays for standard generation; use an explicitly supported keyframe/performance operation for mixed inputs. Must be public/directly fetchable or a Creative Claw URL. In ChatGPT, call import_chatgpt_media for conversation media. Aliases accepted: start_image_url, first_frame_url.'}, 'audio_urls': {'type': 'array', 'items': {'type': 'string'}, 'description': 'Reference audio for reference-to-video. Seedance 2.5 accepts up to 10 clips, each 2â\x80\x9330s and 30s total; Seedance Mini accepts up to 3 clips, each 2â\x80\x9315s and 15s total. H3 Max accepts up to 3 and normalizes Audio 1 syntax automatically for the active provider. Requires at least one reference image or video. For Seedance 2.5 source-video edits, set extras.omni_reference_task_type to edit and preserve the source timeline with duration auto/omitted; for extensions, use extend plus a numeric 4â\x80\x9330s duration.'}, 'image_urls': {'type': 'array', 'items': {'type': 'string'}, 'description': 'Ordered reference images, passed at top level. Use image_urls, including a one-item array, whenever the supplied image is a reference and must not become the literal first frame. Do not combine with image_url or last_frame_url for standard generation. Order defines reference tokens and is never changed. Seedance 2.5 accepts up to 30 and cites @Image1; Seedance Mini accepts up to 9. H3 Max and H3 Max Turbo accept up to 9 and normalize Image 1 syntax for the provider. Call get_model_params for model limits and exceptions.'}, 'resolution': {'enum': ['360p', '480p', '480P', '540p', '720p', '768P', '1080p', '1080P', 'true_1080p', '1440p', '2160p', '2K', '4k'], 'type': 'string', 'description': 'Requested output resolution when supported. Values are model-specific; use get_model_params to discover them. Gemini Omni accepts "360p", "720p", "1080p", or "4k"; MiniMax H3 Max accepts "480P", "768P", or "1080P"; standard H3 accepts "768P" or "2K".'}, 'video_urls': {'type': 'array', 'items': {'type': 'string'}, 'description': "Reference videos for reference-to-video, or one source video for editing. Before editing, compare the source duration with the selected model's uploaded-edit limit. If it is too long, call trim_video for the exact requested time range, wait with check_job, then retry with the completed trimmed URL; do not resend the oversized source. Seedance 2.5 accepts up to 10 clips, each 2â\x80\x9330s and 30s total; Seedance Mini accepts up to 3 clips, each 2â\x80\x9315s and 15s total. H3 Max accepts up to 3 and normalizes Video 1 syntax automatically for the active provider. Other models have their own limits; call get_model_params."}, 'aspect_ratio': {'type': 'string', 'description': 'Aspect ratio (e.g. "16:9", "9:16", "1:1", "4:3")'}, 'character_id': {'type': 'string', 'format': 'uuid', 'pattern': '^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$', 'description': "Select a saved Character (persona). Its description is woven into the prompt and its reference image is used as the start frame (if you don't pass image_url) so the character stays consistent. Use list_characters to find ids."}, 'last_frame_url': {'type': 'string', 'description': 'End frame image URL. Must be public/directly fetchable or returned by import_chatgpt_media/import_media. When provided with image_url, generates a video transitioning from the first frame to the last frame. Supported by Veo 3.1, Kling v3 Pro, MiniMax H3 Max, MiniMax H3, and other compatible models. Alias accepted: end_image_url.'}, 'agentic_prompting': {'type': 'boolean', 'description': "Optional Creative Claw prompt rewrite. Omit it by default to preserve the user's wording while retaining provider-native prompt expansion where supported. Set true only when the user wants a model-specific rewrite. Set false for strict wording and to disable provider-native expansion where supported. Selected Character identity context is appended deterministically either way."}}, 'additionalProperties': {}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'properties': {'estimate_type': {'enum': ['image', 'video', 'speech', '3d_model', 'video_edit', 'html_video'], 'type': 'string', 'description': 'Operation type for cost estimation'}, 'estimate_model': {'type': 'string', 'description': 'Model ID to estimate. Not required when estimate_type is html_video.'}, 'estimate_params': {'type': 'object', 'description': 'Parameters for cost estimation (e.g. width, height, duration)', 'propertyNames': {'type': 'string'}, 'additionalProperties': {}}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'properties': {'product': {'enum': ['1000', '5000', '11000', '27500'], 'type': 'string', 'default': '1000', 'description': 'Credit pack to purchase: "1000" ($10), "5000" ($50), "11000" ($100), or "27500" ($250)'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['id_or_slug'], 'properties': {'id_or_slug': {'type': 'string', 'minLength': 1, 'description': 'Example UUID or slug returned by search_examples.'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['id'], 'properties': {'id': {'type': 'string', 'format': 'uuid', 'pattern': '^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$'}, 'ui_refresh': {'type': 'boolean', 'const': True, 'description': 'Internal film-preview auto-refresh marker.'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['model'], 'properties': {'model': {'type': 'string', 'description': 'Model ID (e.g. "image/nano-banana-2", "video/veo-3.1", "speech/elevenlabs-v3", "sfx/elevenlabs-sound-v2")'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'properties': {'name': {'type': 'string', 'description': 'Theme name. If omitted, returns the default theme.'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['content_type', 'type'], 'properties': {'name': {'type': 'string', 'description': 'Optional name for the asset'}, 'tags': {'type': 'array', 'items': {'type': 'string'}, 'description': 'Optional tags for the asset'}, 'type': {'enum': ['image', 'video', 'audio', '3d_model', 'font', 'zip', 'project'], 'type': 'string', 'description': 'Asset type. Use zip for an ordinary ZIP asset; project is a render_html_video project ZIP.'}, 'purpose': {'enum': ['media', 'voice_clone'], 'type': 'string', 'description': 'Use voice_clone for private voice samples. Returns an asset ID for clone_voice, never a public audio URL.'}, 'filename': {'type': 'string', 'description': 'Original filename. Used to pick the file extension when the browser reports a generic content_type (e.g. application/octet-stream for woff2).'}, 'description': {'type': 'string', 'description': 'Optional description of the asset'}, 'content_type': {'type': 'string', 'description': 'MIME type of the file to upload (e.g. "video/mp4", "image/png", "font/woff2")'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'properties': {'context': {'type': 'string', 'description': 'Optional context about what the user wants to upload or do with the files'}, 'purpose': {'enum': ['media', 'voice_clone'], 'type': 'string', 'description': 'Use voice_clone to upload a private voice sample. The picker returns audio_asset_id for clone_voice instead of a public URL.'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['audio_url'], 'properties': {'audio_url': {'type': 'string', 'format': 'uri', 'description': 'Public URL of the audio file to clean. Strips background noise, music, and reverb to leave only the voice. In ChatGPT, call import_chatgpt_media for an audio file already attached or pasted; call import_media only to open the upload picker.'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'properties': {}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'properties': {}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'properties': {'query': {'type': 'string', 'description': "Free-text search (e.g. 'flux', 'veo', 'fast', 'cheap')"}, 'category': {'enum': ['image', 'video', 'speech', 'audio'], 'type': 'string', 'description': 'Filter by model category. Omit to show all categories.'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'properties': {'include_html': {'type': 'boolean', 'default': False, 'description': 'If true, each result includes the full template HTML. Off by default to keep responses lean â\x80\x94 templates can be large.'}, 'name_contains': {'type': 'string', 'description': 'Case-insensitive substring filter applied to template name and description. Omit to return all templates.'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'properties': {}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['url'], 'properties': {'url': {'type': 'string', 'format': 'uri', 'description': 'URL of the image to load and view'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'properties': {'id': {'type': 'string', 'format': 'uuid', 'pattern': '^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$', 'description': 'Character ID. Required to update; omit to create.'}, 'title': {'type': 'string', 'description': 'Character name. Required when creating.'}, 'image_url': {'type': 'string', 'format': 'uri', 'description': 'Reference portrait / character sheet URL â\x80\x94 the visual anchor for generate_image / generate_video.'}, 'description': {'type': 'string', 'description': 'Appearance, personality, role â\x80\x94 baked into prompts at generation time.'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['action', 'payload'], 'properties': {'action': {'type': 'string', 'maxLength': 64, 'description': 'Supported actions: submit_video_render, set_video_render_mode, get_generation_timing, notify_job, track_ui_event.'}, 'payload': {'type': 'object', 'description': 'Action-specific payload. submit_video_render requires request_id, expected_version, and prompt, and accepts generate_audio when that setting is editable. set_video_render_mode requires mode. get_generation_timing accepts the Chromium HTML render dimensions. notify_job requires job_id.', 'propertyNames': {'type': 'string'}, 'additionalProperties': {}}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['operation'], 'properties': {'audio_url': {'type': 'string', 'description': 'URL of the audio file (for merge_audio_video)'}, 'operation': {'enum': ['merge_audio_video', 'merge_videos', 'merge_audios'], 'type': 'string', 'description': 'The merge operation. merge_audio_video: combine a video with an audio track. merge_videos: concatenate multiple videos. merge_audios: concatenate multiple audio files.'}, 'pad_color': {'enum': ['black', 'white', 'gray'], 'type': 'string', 'default': 'black', 'description': 'Padding color when video_fit is pad. Defaults to black.'}, 'video_fit': {'enum': ['auto', 'crop', 'pad', 'strict'], 'type': 'string', 'default': 'auto', 'description': 'How merge_videos handles aspect-ratio mismatches. auto (default) center-crops to fill the selected canvas; crop explicitly does the same; pad preserves the full frame with letterboxing/pillarboxing; strict rejects mismatches. Videos are never stretched.'}, 'video_url': {'type': 'string', 'description': 'URL of the video file (for merge_audio_video)'}, 'audio_urls': {'type': 'array', 'items': {'type': 'string'}, 'description': 'Audio URLs to concatenate in order (for merge_audios). A single job merges at most 5 inputs. If more are supplied, only the first 5 are submitted; after completion, call merge_media again with the returned merged audio first followed by the remaining URLs.'}, 'target_fps': {'type': 'number', 'maximum': 60, 'minimum': 1, 'description': "Output frame rate for merge_videos, from 1 to 60 FPS. Defaults to the lowest input frame rate, matching fal's behavior."}, 'video_urls': {'type': 'array', 'items': {'type': 'string'}, 'maxItems': 25, 'description': 'Array of video URLs to concatenate in order (for merge_videos), with at most 25 per call. For more, merge the first 25, then merge that result with the remaining videos.'}, 'audio_format': {'enum': ['mp3', 'm4a', 'wav'], 'type': 'string', 'description': 'Output format for merge_audios. When omitted, WAV-only inputs remain WAV, M4A-only inputs remain M4A, and other combinations use MP3.'}, 'start_offset': {'type': 'number', 'minimum': 0, 'description': 'Seconds after the video starts when the replacement audio should begin (for merge_audio_video). Defaults to 0.'}, 'canvas_video_index': {'type': 'integer', 'default': 0, 'maximum': 9007199254740991, 'minimum': 0, 'description': 'Zero-based video index whose width, height, and aspect ratio define the output canvas for merge_videos. Defaults to 0 (the first video).'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['media_url', 'type'], 'properties': {'type': {'enum': ['image', 'video'], 'type': 'string', 'description': 'Type of media: image or video'}, 'force_new': {'type': 'boolean', 'description': 'Bypass the 10-minute same-workspace duplicate guard. Set true only when the user explicitly wants another paid result from identical input.'}, 'media_url': {'type': 'string', 'description': 'URL of the image or video to remove background from'}, 'output_format': {'enum': ['webm', 'mp4'], 'type': 'string', 'description': 'Output format (video only). webm supports true alpha transparency. mp4 requires a background_color or produces black background. Default: webm'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['html'], 'properties': {'html': {'type': 'string', 'description': "HTML markup to render. Rendered via headless Chromium â\x80\x94 full CSS surface is available (flexbox, grid, filter, mask-image, transform, <style> blocks, class selectors, pseudo-elements, variable fonts). Write HTML the way you would for a real browser. Use <img src='{{token}}'> for any image passed via inline_images."}, 'name': {'type': 'string', 'description': 'Optional asset name for the resulting image'}, 'tags': {'type': 'array', 'items': {'type': 'string'}, 'description': 'Optional tags to attach to the saved asset'}, 'width': {'type': 'integer', 'default': 1200, 'maximum': 9007199254740991, 'description': 'Output width in pixels', 'exclusiveMinimum': 0}, 'height': {'type': 'integer', 'default': 630, 'maximum': 9007199254740991, 'description': 'Output height in pixels', 'exclusiveMinimum': 0}, 'inline_images': {'type': 'array', 'items': {'type': 'object', 'required': ['token', 'url'], 'properties': {'url': {'type': 'string'}, 'token': {'type': 'string'}}}, 'description': 'Images to embed. Each {token, url} pair: {{token}} in the HTML is substituted with the raw URL, and Chromium fetches the image directly at render time. URLs must be publicly reachable.'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'properties': {'fps': {'type': 'integer', 'default': 30, 'maximum': 9007199254740991, 'description': 'Frames per second â\x80\x94 24 (cinematic), 30 (standard), or 60 (smooth). Default: 30', 'exclusiveMinimum': 0}, 'html': {'type': 'string', 'minLength': 1, 'description': 'HTML composition to render as a video. Rendered via HyperFrames + headless Chromium on Modal. Supports GSAP animations, CSS transitions/keyframes, and Tailwind CSS utility classes. Supply one complete HTML document with a fixed-size root declaring data-composition-id, data-width, data-height and data-duration matching the requested output. Use data-start/data-duration for clips and register a paused GSAP timeline as window.__timelines[compositionId]. For canvas/WebGL, redraw from an absolute-time property setter tween (ease: none, lazy: false), not an onUpdate callback. Alternatively supply project_url for any publicly downloadable project ZIP, or project_asset_id for a private workspace ZIP. For sound, add an <audio id="..." src="https://..." data-start="0" data-duration="..."> element. Audio src must be an absolute HTTP(S) URL; inline data/blob URLs and relative paths are unsupported.'}, 'name': {'type': 'string', 'description': 'Optional asset name for the saved video'}, 'tags': {'type': 'array', 'items': {'type': 'string'}, 'description': 'Optional tags to attach to the saved asset'}, 'width': {'type': 'integer', 'default': 1920, 'maximum': 7680, 'minimum': 2, 'description': 'Output width in pixels (even number, 2â\x80\x937680)'}, 'format': {'enum': ['mp4', 'webm', 'mov'], 'type': 'string', 'default': 'mp4', 'description': 'Output video format: mp4 (default), webm, or mov'}, 'height': {'type': 'integer', 'default': 1080, 'maximum': 7680, 'minimum': 2, 'description': 'Output height in pixels (even number, 2â\x80\x937680)'}, 'duration': {'type': 'number', 'default': 5, 'maximum': 300, 'description': 'Video duration in seconds (maximum: 300; default: 5)', 'exclusiveMinimum': 0}, 'project_url': {'type': 'string', 'format': 'uri', 'description': 'Publicly downloadable HTTPS URL for a HyperFrames project ZIP. Use this directly for ZIP URLs returned by get_example. The URL may be hosted on any public domain and may include temporary query parameters.'}, 'project_asset_id': {'type': 'string', 'format': 'uuid', 'pattern': '^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$', 'description': 'Workspace ZIP asset ID, instead of html. Both zip and project asset types are accepted; project is recommended for dedicated render projects. package.json auto-detects npm; a build script must produce dist/index.html, otherwise use index.html.'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['renders'], 'properties': {'name': {'type': 'string', 'description': 'Optional asset name prefix for the rendered outputs. With multiple variants, each asset is suffixed with its index.'}, 'tags': {'type': 'array', 'items': {'type': 'string'}, 'description': 'Optional tags to attach to every saved asset'}, 'renders': {'type': 'array', 'items': {'type': 'object', 'properties': {'modifications': {'type': 'object', 'default': {}, 'description': 'Values to substitute into the template for this variant. Keys must match parameter names defined at create_template time.', 'propertyNames': {'type': 'string'}, 'additionalProperties': {'anyOf': [{'type': 'string'}, {'type': 'number'}]}}}}, 'maxItems': 4, 'minItems': 1, 'description': 'Variants to render in one call (1..4). Each entry is the per-variant params (modifications). All variants share the same template and output settings. Example: 3 versions of a card with different headlines â\x86\x92 3 entries with different modifications.'}, 'size_names': {'type': 'array', 'items': {'type': 'string'}, 'description': 'Which named sizes to render at. Omit to render only the default size. Pass ["all"] to render every configured size. Pass specific names (e.g. ["story", "square"]) to render a subset. Each (variant Ã\x97 size) is charged independently.'}, 'template_id': {'type': 'string', 'format': 'uuid', 'pattern': '^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$', 'description': 'Template ID returned by create_template. Provide either this or template_name.'}, 'template_name': {'type': 'string', 'description': 'Template name to look up (case-insensitive). Use list_templates to discover names. Provide either this or template_id.'}, 'width_override': {'type': 'integer', 'maximum': 9007199254740991, 'description': "Override the template's stored width (single-size renders only). Applies to all variants.", 'exclusiveMinimum': 0}, 'height_override': {'type': 'integer', 'maximum': 9007199254740991, 'description': "Override the template's stored height (single-size renders only). Applies to all variants.", 'exclusiveMinimum': 0}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['video_url'], 'properties': {'mode': {'enum': ['stretch', 'pad', 'crop'], 'type': 'string', 'description': 'Scaling mode. stretch=distort to fit, pad=letterbox with color, crop=center-crop. Default: stretch'}, 'width': {'type': 'integer', 'maximum': 7680, 'minimum': 2, 'description': 'Target width in pixels (2-7680). Auto-calculates height if only width provided'}, 'height': {'type': 'integer', 'maximum': 4320, 'minimum': 2, 'description': 'Target height in pixels (2-4320). Auto-calculates width if only height provided'}, 'pad_color': {'enum': ['black', 'white', 'red', 'green', 'blue', 'gray'], 'type': 'string', 'description': "Padding color when mode is 'pad'. Default: black"}, 'video_url': {'type': 'string', 'description': 'URL of the video to scale'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'properties': {'name': {'type': 'string', 'description': 'Filter by exact asset name'}, 'tags': {'type': 'array', 'items': {'type': 'string'}, 'description': 'Filter by tags (returns assets matching ANY of the given tags)'}, 'type': {'enum': ['image', 'video', 'audio', '3d_model', 'document', 'zip'], 'type': 'string', 'description': 'Filter by media type'}, 'limit': {'type': 'number', 'default': 20, 'maximum': 100, 'minimum': 1, 'description': 'Maximum number of results to return (default 20, max 100)'}, 'query': {'type': 'string', 'description': 'Search string to match against the description, asset name, or generation prompt'}, 'offset': {'type': 'number', 'default': 0, 'minimum': 0, 'description': 'Number of results to skip for pagination'}, 'source': {'enum': ['generated', 'uploaded', 'imported'], 'type': 'string', 'description': 'Filter by asset source'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'properties': {'tags': {'type': 'array', 'items': {'type': 'string'}, 'maxItems': 10, 'description': 'Optional tags that every result must contain.'}, 'limit': {'type': 'integer', 'default': 30, 'maximum': 50, 'minimum': 1, 'description': 'Results per page.'}, 'query': {'type': 'string', 'description': 'What the user wants to make, including subject, style, mood, or use case.'}, 'cursor': {'type': 'string', 'description': 'Opaque cursor from a previous result page.'}, 'model_id': {'type': 'string', 'description': 'Optional exact Creative Claw model ID. Omit when the user has not chosen a model.'}, 'output_type': {'enum': ['image', 'video', 'audio'], 'type': 'string', 'description': 'Optional output filter. Omit to search every media type.'}, 'render_type': {'enum': ['html_video'], 'type': 'string', 'description': 'Find executable HyperFrames HTML-video examples, either single HTML or a ZIP project. Output remains video.'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['message', 'category', 'source'], 'properties': {'source': {'enum': ['agent', 'user'], 'type': 'string', 'description': 'Who the feedback comes from. "agent" = YOU noticed friction while working (a missing tool, a confusing error, output that needed rework). "user" = you are relaying the user\'s own words â\x80\x94 quote them.'}, 'message': {'type': 'string', 'minLength': 1, 'description': 'Specific feedback: what worked well, broke, was missing, or was confusing.'}, 'category': {'enum': ['bug', 'missing_feature', 'confusing', 'praise', 'other'], 'type': 'string', 'description': "bug = something errored or produced wrong output; missing_feature = a capability the user wanted that doesn't exist; confusing = a tool/flow was hard to use or its output was unclear; praise = positive feedback worth recording; other = anything else."}, 'toolName': {'type': 'string', 'description': 'The tool involved, if any (e.g. "generate_video", "render_template").'}, 'relatedIds': {'type': 'array', 'items': {'type': 'string'}, 'description': 'Related job or asset IDs, if known. Saved as supplied, even if invalid.'}, 'attemptedTask': {'type': 'string', 'description': 'What the user was trying to accomplish.'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'properties': {'audio_url': {'type': 'string', 'format': 'uri', 'description': 'Public URL of the audio file to transcribe. Supported formats: mp3, ogg, wav, m4a, aac. Pass exactly one of audio_url or video_url. Audio is preferred when you have it â\x80\x94 much smaller payloads, faster jobs, cheaper.'}, 'video_url': {'type': 'string', 'format': 'uri', 'description': 'Public URL of a video file or public video-hosting URL. Direct Scribe supports hosted audio/video, YouTube, TikTok, Instagram, and other video-hosting URLs when the provider can fetch them. Pass exactly one of audio_url or video_url.'}, 'language_code': {'type': 'string', 'description': 'ISO-639 language code (e.g. "en", "he", "es"). Locale tags such as "en-GB" are accepted and normalized to their base language. Omit to auto-detect.'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'properties': {'duration': {'type': 'number', 'description': 'Duration in seconds from start_time. Ignored if end_time is provided. Default: 2'}, 'end_time': {'type': 'number', 'description': 'End time in seconds. If supplied, it takes precedence over duration'}, 'audio_url': {'type': 'string', 'description': 'Public URL of the audio to trim. Pass exactly one of audio_url or video_url. Audio trimming runs asynchronously; use check_job with the returned job ID.'}, 'video_url': {'type': 'string', 'description': 'Public URL of the video to trim. Pass exactly one of video_url or audio_url. Existing video trimming behavior and output remain unchanged.'}, 'start_time': {'type': 'number', 'description': 'Start time in seconds. Default: 1'}, 'audio_format': {'enum': ['mp3', 'm4a', 'wav'], 'type': 'string', 'description': 'Audio output format when audio_url is supplied. Default: mp3'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['id'], 'properties': {'id': {'type': 'string', 'format': 'uuid', 'pattern': '^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$', 'description': 'The asset ID to update'}, 'name': {'type': 'string', 'description': 'Set a unique name for this asset. Must be unique across your assets. Use null to clear the name.'}, 'tags': {'type': 'array', 'items': {'type': 'string'}, 'description': 'Set tags for this asset (replaces existing tags). Pass an empty array to clear tags.'}, 'clear_name': {'type': 'boolean', 'default': False, 'description': "Set to true to remove the asset's name"}, 'description': {'type': 'string', 'description': 'Set or update the asset description'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['id'], 'properties': {'id': {'type': 'string', 'format': 'uuid', 'pattern': '^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$'}, 'name': {'type': 'string'}, 'brief': {'type': 'string'}, 'shots': {'type': 'array', 'items': {'type': 'object', 'required': ['id', 'description'], 'properties': {'id': {'type': 'string', 'minLength': 1, 'description': 'Stable shot id, for example "s1"'}, 'model': {'type': 'string', 'description': 'Video model id selected for this shot'}, 'prompt': {'type': 'string', 'description': 'Approved motion prompt for the video model'}, 'status': {'enum': ['pending', 'storyboard', 'clip', 'done', 'error'], 'type': 'string'}, 'clipUrl': {'type': 'string', 'format': 'uri', 'description': 'Approved rendered clip URL'}, 'audioUrl': {'type': 'string', 'format': 'uri', 'description': 'Approved per-shot narration or dialogue URL'}, 'durationS': {'type': 'number', 'description': 'Shot duration supported by its model', 'exclusiveMinimum': 0}, 'narration': {'type': 'string', 'description': 'Voiceover or dialogue text for this shot'}, 'description': {'type': 'string', 'minLength': 1, 'description': 'What happens in this shot'}, 'storyboardUrl': {'type': 'string', 'format': 'uri', 'description': 'Approved clean storyboard image URL'}}, 'additionalProperties': False}, 'description': 'Full validated shot list; replaces the existing list.'}, 'status': {'enum': ['drafting', 'script_ok', 'storyboard_ok', 'rendering', 'preview_ok', 'final'], 'type': 'string'}, 'logline': {'type': 'string'}, 'theme_id': {'type': 'string', 'format': 'uuid', 'pattern': '^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$'}, 'audio_url': {'type': 'string', 'format': 'uri', 'description': 'Full narration track URL'}, 'patch_shots': {'type': 'array', 'items': {'type': 'object', 'required': ['id'], 'properties': {'id': {'type': 'string', 'minLength': 1}, 'model': {'type': 'string', 'description': 'Video model id selected for this shot'}, 'prompt': {'type': 'string', 'description': 'Approved motion prompt for the video model'}, 'status': {'enum': ['pending', 'storyboard', 'clip', 'done', 'error'], 'type': 'string'}, 'clipUrl': {'type': 'string', 'format': 'uri', 'description': 'Approved rendered clip URL'}, 'audioUrl': {'type': 'string', 'format': 'uri', 'description': 'Approved per-shot narration or dialogue URL'}, 'durationS': {'type': 'number', 'description': 'Shot duration supported by its model', 'exclusiveMinimum': 0}, 'narration': {'type': 'string', 'description': 'Voiceover or dialogue text for this shot'}, 'description': {'type': 'string', 'minLength': 1, 'description': 'What happens in this shot'}, 'storyboardUrl': {'type': 'string', 'format': 'uri', 'description': 'Approved clean storyboard image URL'}}, 'additionalProperties': False}, 'description': 'Validated partial shot updates merged by stable id.'}, 'assembled_url': {'type': 'string', 'format': 'uri', 'description': 'Final stitched film URL'}, 'character_ids': {'type': 'array', 'items': {'type': 'string', 'format': 'uuid', 'pattern': '^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$'}}, 'target_duration_s': {'type': 'integer', 'maximum': 3600, 'minimum': 4}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'properties': {'html': {'type': 'string', 'description': 'html kind only. Replacement HTML markup.'}, 'name': {'type': 'string', 'minLength': 1, 'description': 'New name (unique per user, case-insensitive)'}, 'sizes': {'type': 'array', 'items': {'type': 'object', 'required': ['name', 'width', 'height'], 'properties': {'name': {'type': 'string', 'pattern': '^[a-zA-Z0-9_-]+$'}, 'label': {'type': 'string'}, 'width': {'type': 'integer', 'maximum': 9007199254740991, 'exclusiveMinimum': 0}, 'height': {'type': 'integer', 'maximum': 9007199254740991, 'exclusiveMinimum': 0}, 'bodyClass': {'type': 'string', 'pattern': '^[\\w\\s-]+$'}, 'isDefault': {'type': 'boolean'}}}, 'description': 'html kind only. Replacement named sizes list. Pass [] to clear and fall back to width/height.'}, 'width': {'type': 'integer', 'maximum': 9007199254740991, 'description': 'html kind only. New width.', 'exclusiveMinimum': 0}, 'height': {'type': 'integer', 'maximum': 9007199254740991, 'description': 'html kind only. New height.', 'exclusiveMinimum': 0}, 'prompt': {'type': 'string', 'description': 'Generative kind only. Replacement prompt body.'}, 'parameters': {'type': 'array', 'items': {'type': 'object', 'required': ['name', 'type'], 'properties': {'name': {'type': 'string', 'pattern': '^[a-zA-Z_][a-zA-Z0-9_]*$'}, 'type': {'enum': ['text', 'image_url', 'color', 'number', 'boolean'], 'type': 'string'}, 'default': {'type': 'string'}, 'required': {'type': 'boolean'}, 'description': {'type': 'string'}}}, 'description': 'Replacement parameter list'}, 'description': {'type': 'string', 'description': 'New description'}, 'template_id': {'type': 'string', 'format': 'uuid', 'pattern': '^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$', 'description': 'Template ID. Provide either this or template_name.'}, 'model_params': {'type': 'object', 'description': 'Generative kind only. Replacement model_params object.', 'propertyNames': {'type': 'string'}, 'additionalProperties': {}}, 'template_name': {'type': 'string', 'description': 'Template name to look up (case-insensitive). Provide either this or template_id.'}, 'recommended_model': {'type': 'string', 'description': 'Generative kind only. New registry model id.'}, 'reference_asset_ids': {'type': 'array', 'items': {'type': 'string', 'format': 'uuid', 'pattern': '^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$'}, 'description': 'Generative kind only. Replacement reference asset id list.'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'properties': {'data': {'type': 'object', 'description': 'Theme data â\x80\x94 colors, fonts, logos, etc. Merged with existing data by default. Pass only when updating data; pass only `images` to update brand reference images.', 'propertyNames': {'type': 'string'}, 'additionalProperties': {}}, 'name': {'type': 'string', 'description': "Theme name. Creates the theme if it doesn't exist. Optional only when interactive=true (the user picks/creates a theme inside the editor)."}, 'images': {'type': 'array', 'items': {'type': 'string', 'format': 'uri'}, 'description': "Brand reference images for this theme â\x80\x94 passed to image/video models as visual references. Replaces the entire images array when provided. The first image is automatically saved as the theme's thumbnail."}, 'override': {'type': 'boolean', 'default': False, 'description': 'If true, replaces the entire theme data. If false (default), shallow-merges with existing data â\x80\x94 top-level keys are replaced, missing keys are preserved. Only applies to `data`; `images` is always replaced.'}, 'interactive': {'type': 'boolean', 'description': "When true, open the visual theme editor (fonts, colors, logos, reference image) in the chat instead of mutating directly. The user fills in the form and clicks save; the editor calls update_theme again with interactive=false to persist. Use this whenever the user says 'edit my theme', 'create a theme', 'help me set up a brand', etc."}, 'set_default': {'type': 'boolean', 'default': False, 'description': 'If true, makes this theme the default.'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['url', 'content_type'], 'properties': {'url': {'type': 'string', 'format': 'uri', 'description': 'Direct URL of a publicly accessible image, video, audio, 3D-model, or ZIP file to import and store permanently. Webpage and JSON API URLs are rejected.'}, 'name': {'type': 'string', 'description': 'Optional unique name for the asset'}, 'tags': {'type': 'array', 'items': {'type': 'string'}, 'description': 'Optional tags for the asset'}, 'type': {'enum': ['image', 'video', 'audio', '3d_model', 'zip'], 'type': 'string', 'description': 'Asset type. Optional when content_type clearly identifies an image, video, audio file, 3D model, or ZIP archive.'}, 'description': {'type': 'string', 'description': 'Optional description of the asset'}, 'content_type': {'type': 'string', 'description': 'MIME type of the file (e.g. "image/png", "video/mp4", "audio/mp3")'}}}
入力スキーマ
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['media_url', 'type'], 'properties': {'seed': {'type': 'number', 'description': 'Seed for reproducibility (image, clarity model only)'}, 'type': {'enum': ['image', 'video'], 'type': 'string', 'description': 'Type of media: image or video'}, 'prompt': {'type': 'string', 'description': 'Guide the upscaling (image, clarity model only). Describes desired output quality.'}, 'media_url': {'type': 'string', 'description': 'URL of the image or video to upscale'}, 'creativity': {'type': 'number', 'maximum': 1, 'minimum': 0, 'description': 'Creativity level 0-1 (image, clarity model only). Higher = more hallucinated detail. Default: 0'}, 'target_fps': {'enum': ['30', '60'], 'type': 'string', 'description': 'Target FPS (video, bytedance model only). Default: 30'}, 'resemblance': {'type': 'number', 'maximum': 1, 'minimum': 0, 'description': 'Resemblance to original 0-1 (image, clarity model only). Higher = closer to input. Default: 1'}, 'upscale_model': {'type': 'string', 'description': 'Upscaler model. For images: aura (cheapest/fastest, 4x fixed), clarity (best quality, 1-4x, prompt-guided), recraft (crisp/simple). For videos: realesrgan (cheapest, 1-8x), bytedance (higher quality with resolution presets). Defaults: aura for images, realesrgan for videos.'}, 'upscale_scale': {'type': 'integer', 'maximum': 8, 'minimum': 1, 'description': 'Scale factor 1-8x (video, realesrgan model only). Default: 2'}, 'upscale_factor': {'type': 'number', 'maximum': 4, 'minimum': 1, 'description': 'Upscale factor 1-4x (image, clarity model only). Aura is fixed at 4x. Default: 2'}, 'target_resolution': {'enum': ['1080p', '2k', '4k'], 'type': 'string', 'description': 'Target resolution (video, bytedance model only). Default: 1080p'}, 'enhancement_preset': {'enum': ['general', 'ugc', 'short_series', 'aigc', 'old_film'], 'type': 'string', 'description': 'Enhancement preset (video, bytedance model only). Default: general'}}}
最近のツール変更
類似のMCPサーバー
MiOffice — AI-Powered Workspace Studio
Provides browser-based tools for processing PDFs, images, video, and audio, including generation, enhancement, conversion, transc…
BlitzReels Video Editor
Creates and edits short-form videos, including timelines, captions, transitions, AI-generated visuals, music, voiceovers, sound e…
Morpha
Enables creating, animating, and exporting layered short-form video projects.
switch
Enables management and exploration of an account-scoped library of AI-generated images and videos.
Magnific
Provides image enhancement and design tools, AI image and audio generation, text-to-speech, creation management, and reusable cre…
Uwear
Supports AI fashion photoshoot production with garments, outfits, avatars, locations, art direction, image generation, backdrops,…
aicut
Generates and analyzes AI videos, images, audio, talking-avatar clips, scripted stories, and social posts, with staged review and…
RegiAI
Generates and edits images, videos, audio, music, speech, subtitles, avatars, and related media asynchronously.