> ## Documentation Index > Fetch the complete documentation index at: https://tryunsora.com/docs/llms.txt > Use this file to discover all available pages before exploring further. # Create clipping > Submit a video for AI clipping. By default the AI auto-detects the most viral clips. Set query to switch to Find Moments mode: describe the moments you want in natural language (e.g. "funny reactions") and the AI extracts matching clips instead — targetDuration is ignored in that mode. Cost is charged per started minute of the source video either way. ## OpenAPI ````yaml POST /clippings/create openapi: 3.1.0 info: title: Unsora API version: 1.2.0 description: >- Public API for Unsora integrations. Authenticate with Clerk JWT or uns_live_* API key. API keys are rate-limited to 60 requests/minute and support idempotency keys. Async jobs complete via polling GET /image/status/{id}. servers: - url: https://mvp.tryunsora.com/api/v1 description: Versioned API (recommended) security: [] tags: - name: Clippings - name: Connect - name: Image Generations - name: Music Generations - name: Influencer Studio - name: Posts - name: Thumbnails - name: User - name: Video Generations paths: /clippings/create: post: tags: - Clippings summary: Create AI clipping job description: >- Submit a video for AI clipping. By default the AI auto-detects the most viral clips. Set query to switch to Find Moments mode: describe the moments you want in natural language (e.g. "funny reactions") and the AI extracts matching clips instead — targetDuration is ignored in that mode. Cost is charged per started minute of the source video either way. requestBody: required: true content: application/json: schema: type: object properties: videoUrl: type: string description: >- Public URL of the source video (YouTube link or direct video URL). example: https://www.youtube.com/watch?v=dQw4w9WgXcQ sourceLang: type: string nullable: true description: >- Source language code of the video (e.g. en, es). Omit or null to auto-detect. targetLang: type: string nullable: true description: >- Target language for clip titles, descriptions, and captions. Omit or null to keep the source language. targetDuration: type: string description: >- Desired length range for each clip. Defaults to auto (up to 90s). enum: - auto - lt30 - 30-60 - 60-90 - 90-3min - gt3min default: auto query: type: string maxLength: 500 nullable: true description: >- Find Moments mode: natural-language description of the moments to extract (e.g. "funny reactions", "product demos", "goal moments and key plays"). When set, targetDuration is ignored. Omit to auto-detect the most viral clips. example: funny reactions limit: type: integer minimum: 1 maximum: 20 nullable: true description: >- Maximum number of clips to return, ranked by virality score (highest first). Omit to return all viral-worthy clips. ratio: type: string enum: - '9:16' - '1:1' - '4:5' - '16:9' - original default: original description: >- Output aspect ratio. AI reframe keeps the main subject centered when cropping. Use 9:16 for TikTok/Reels/Shorts, 1:1 or 4:5 for feed posts, 16:9 for YouTube. Omit or use original to keep the source ratio. enableCaption: type: boolean default: false description: Burn animated captions into the exported clips. captionStyle: type: string description: >- Caption style preset, named by look — the color is the accent on the spoken word. glow-* styles add a soft glow, static-* styles don't animate, gaming-* are bold streamer styles, white-card/black-box place text on a card. Only used when enableCaption is true. Defaults to classic-yellow. enum: - classic-yellow - white-card - black-box - bold-white - white-underline - playful-green - sketch-blue - bubble-blue - bubble-green - glow-yellow - duo-green - duo-purple - elegant-purple - neon-cyan - glow-pink - bold-yellow - soft-orange - glow-green - glow-orange - retro-outline - mint-bubble - comic-duo - static-outline - static-comic - static-minimal - static-green - static-yellow - static-orange - gaming-magenta - gaming-green - gaming-yellow - gaming-white - gaming-purple-box - gaming-cyan - gaming-orange-box default: classic-yellow required: - videoUrl responses: '201': description: >- Job queued. Clips render asynchronously — poll GET /clippings/status/{clippingId} until status is COMPLETED. content: application/json: schema: $ref: '#/components/schemas/ClippingCreateResponse' example: success: true data: id: cmcl0x2ab0001abcd1234efgh status: QUEUED videoUrl: https://www.youtube.com/watch?v=dQw4w9WgXcQ config: sourceLang: null targetLang: null targetDuration: DURATION_30_60 limit: 5 enableCaption: true captionStyle: temp-7 ratio: RATIO_9_16 createdAt: '2026-06-10T12:00:00.000Z' clips: [] creditsUsed: 40 creditsDeducted: 40 creditsRemaining: 960 '400': description: >- Invalid input (bad videoUrl, targetDuration, limit, ratio, or captionStyle), or video length could not be determined. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' example: success: false error: Invalid ratio. Use 9:16, 1:1, 4:5, 16:9, or original '402': description: >- Insufficient credits. Cost is charged per started minute of the source video. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' example: success: false error: Insufficient credits. Need 40, have 10 creditsRemaining: 10 security: - bearerAuth: [] components: schemas: ClippingCreateResponse: type: object properties: success: type: boolean data: $ref: '#/components/schemas/ClippingJob' creditsDeducted: type: integer creditsRemaining: type: integer required: - success - data ErrorResponse: type: object properties: success: type: boolean error: type: string code: type: string message: type: string required: - error ClippingJob: type: object additionalProperties: true properties: id: type: string status: type: string enum: - QUEUED - PROCESSING - COMPLETED - FAILED videoUrl: type: string error: anyOf: - type: string - type: 'null' createdAt: type: string format: date-time updatedAt: type: string format: date-time clips: type: array items: $ref: '#/components/schemas/ClippingClip' config: $ref: '#/components/schemas/ClippingConfig' creditsUsed: type: integer required: - id - status ClippingClip: type: object additionalProperties: true properties: id: type: string order: type: integer description: Clip index, ranked by virality (0 = highest). title: type: string description: AI-generated clip title. startTime: type: number description: Start offset in the source video, seconds. endTime: type: number description: End offset in the source video, seconds. duration: type: number description: Clip length in seconds. status: type: string outputAsset: type: object nullable: true additionalProperties: true description: >- Rendered clip video. Contains the downloadable url once the job completes. thumbnailAsset: type: object nullable: true additionalProperties: true description: Clip thumbnail image. metadata: type: object additionalProperties: true description: >- Extra AI output: tags, desc (suggested description), score (virality). ClippingConfig: type: object description: >- Job settings echoed back in resolved form: targetDuration as DURATION_*, ratio as RATIO_* (null when keeping the original ratio), captionStyle as the underlying template ID. properties: sourceLang: type: string nullable: true targetLang: type: string nullable: true targetDuration: type: string enum: - DURATION_0_30 - DURATION_0_90 - DURATION_30_60 - DURATION_60_90 - DURATION_90_180 - DURATION_180_300 query: type: string nullable: true description: >- Find Moments query the job was created with; null for regular viral-clip jobs. limit: type: integer nullable: true ratio: type: string nullable: true enum: - RATIO_9_16 - RATIO_1_1 - RATIO_4_5 - RATIO_16_9 - null enableCaption: type: boolean captionStyle: type: string nullable: true securitySchemes: bearerAuth: type: http scheme: bearer description: Clerk session JWT or Unsora API key (uns_live_*) ````