Weftly

{
  "name": "Weftly",
  "description": "Upload audio or video for transcription or summarization. No account required. Two payment rails: Stripe SPT (fiat) and Tempo crypto (USDC). Files auto-delete after 4-24 hours.",
  "url": "https://api.weftly.ai",
  "version": "1.0.0",
  "provider": {
    "organization": "Woven Record Media",
    "url": "https://weftly.ai"
  },
  "capabilities": {
    "streaming": true,
    "pushNotifications": false
  },
  "skills": [
    {
      "id": "transcribe-audio",
      "name": "Transcribe Audio",
      "description": "Transcribe an audio file (MP3, WAV, M4A, OGG, FLAC, WebM, AAC) with timestamps. Output formats: SRT, TXT, VTT, JSON, words (per-word timestamps, schema at /.well-known/weftly-transcript-v2.schema.json). Cost: $0.50.",
      "tags": [
        "transcription",
        "audio",
        "srt",
        "vtt",
        "txt",
        "json",
        "words",
        "word-level",
        "speech-to-text"
      ],
      "examples": [
        "Transcribe this podcast episode",
        "Convert audio recording to text with timestamps",
        "Get word-level timestamps for clip extraction"
      ]
    },
    {
      "id": "transcribe-video",
      "name": "Transcribe Video",
      "description": "Transcribe a video file (MP4, MOV, WebM, MKV, AVI) with timestamps. Includes audio extraction. Output formats: SRT, TXT, VTT, JSON, words (per-word timestamps, schema at /.well-known/weftly-transcript-v2.schema.json) \u2014 words format supports clipping, multicamera edits, and edit-video-from-transcript. Cost: $1.00.",
      "tags": [
        "transcription",
        "video",
        "srt",
        "vtt",
        "txt",
        "json",
        "words",
        "word-level",
        "speech-to-text",
        "subtitles",
        "video-editing"
      ],
      "examples": [
        "Transcribe this video interview",
        "Generate subtitles for this MP4",
        "Get word-level timestamps for editing this video on word boundaries"
      ]
    },
    {
      "id": "summarize-audio",
      "name": "Summarize Audio",
      "description": "Transcribe and summarize an audio file. Returns transcript (SRT, TXT, VTT, JSON, or words with per-word timestamps) and text summary. Cost: $0.75.",
      "tags": [
        "summarization",
        "audio",
        "speech-to-text",
        "words",
        "word-level"
      ],
      "examples": [
        "Summarize this podcast episode",
        "Give me key points from this audio recording"
      ]
    },
    {
      "id": "summarize-video",
      "name": "Summarize Video",
      "description": "Transcribe and summarize a video file. Returns transcript (SRT, TXT, VTT, JSON, or words with per-word timestamps) and text summary. Includes audio extraction. Cost: $1.25.",
      "tags": [
        "summarization",
        "video",
        "speech-to-text",
        "words",
        "word-level"
      ],
      "examples": [
        "Summarize this video interview",
        "Give me key points from this video"
      ]
    },
    {
      "id": "find-clips-video",
      "name": "Find Clips",
      "description": "Identify ranked candidate clips in a video. Transcribes the source, then has Claude analyze the transcript to surface compelling 2\u20135 minute moments with titles, hooks, narrative summaries, scoring, and the FULL TEXT of each candidate so callers can preview content before paying for extract_clip. Optional `query` parameter switches from \"best clips\" discovery mode to query mode (e.g., \"they discuss pricing\"). Output schema at /.well-known/weftly-clips-v1.schema.json. Also returns the SRT and word-level transcript as free byproducts. Cost: $2.00.",
      "tags": [
        "clipping",
        "video-analysis",
        "highlights",
        "discovery",
        "words",
        "word-level",
        "video-editing"
      ],
      "examples": [
        "Find the best clips in this interview",
        "Find clips where they discuss pricing in this call",
        "Identify shareable moments from this podcast"
      ]
    },
    {
      "id": "extract-clip",
      "name": "Extract Clip",
      "description": "Cut and assemble a clip from a prior video job (find_clips, summarize, or video transcribe). Pass one segment for a simple cut, or multiple non-contiguous segments to compose a single highlight reel \u2014 same flat $0.50 either way. Frame-accurate boundaries, 15ms audio fades at segment joins. Outputs clip.mp4 plus, by default, clipped SRT + word-level transcript stitched to match the assembled output. Total output duration capped at 30 minutes per call (margin protection). Cost: $0.50, paid via MPP \u2014 accepts Tempo USDC and Stripe SPT. The challenge's WWW-Authenticate header and /.well-known/mpp.json are authoritative for the supported methods.",
      "tags": [
        "clipping",
        "video-editing",
        "extraction",
        "compose",
        "highlights"
      ],
      "examples": [
        "Extract clip 1 from yesterday's find_clips job",
        "Combine clips 1 and 3 from job J1 into one highlight reel",
        "Cut from 12:34 to 15:58 of the call I just transcribed"
      ]
    },
    {
      "id": "extract-vertical-clip",
      "name": "Extract Vertical Clip",
      "description": "Cut a 9:16 vertical clip from any prior video job (find_clips, summarize, or video transcribe), suitable for direct upload to TikTok, Instagram Reels, or YouTube Shorts. 1080\u00d71920 H.264/AAC mp4, center-cropped from a horizontal source. Single segment only; max output duration depends on the profile (240s for tiktok-primary/tiktok-primary-720p, 180s for instagram-reels, 60s for instagram-stories). Pair with find_clips to pick a moment, then call this for a download-ready vertical mp4 in under 5 minutes. Cost: $0.50, paid via MPP \u2014 accepts Tempo USDC and Stripe SPT. The challenge's WWW-Authenticate header and /.well-known/mpp.json are authoritative for the supported methods.",
      "tags": [
        "clipping",
        "video-editing",
        "vertical",
        "tiktok",
        "reels",
        "shorts",
        "social"
      ],
      "examples": [
        "Make clip 1 from yesterday's find_clips job into a TikTok",
        "Vertical clip from 12:34 to 13:30 of the interview",
        "Cut the best moment as a Reel for Instagram"
      ]
    },
    {
      "id": "publish-youtube",
      "name": "Publish to YouTube",
      "description": "Publish a video to your YouTube channel from a prior transcribe or summarize job (referenced by source_job_id) \u2014 no re-upload. Attaches the SRT transcript as captions and, for summarize parents, the summary as the video description. Runs as a durable workflow (handles large files, refresh tokens, chunk-upload retries). Requires a connected YouTube account via Google OAuth (one-time consent; the Weftly app is Google-verified for the youtube.upload scope). Cost: $1.75, paid via MPP \u2014 accepts Tempo USDC and Stripe SPT.",
      "tags": [
        "youtube",
        "publishing",
        "video",
        "captions",
        "oauth"
      ],
      "examples": [
        "Publish my summarized interview to YouTube",
        "Upload this transcribed talk to YouTube with captions, unlisted"
      ]
    },
    {
      "id": "clips-vertical",
      "name": "Clip Pack, Vertical",
      "description": "Cut up to 5 vertical 9:16 clips (1080\u00d71920, TikTok/Reels/Shorts-ready) from a prior find_clips job in one batch purchase \u2014 you choose which candidates to cut, each with an optional title, delivered as downloadable files. One purchase, one job id \u2014 no separate extract_vertical_clip call per clip to track. No publishing in v1. Cost: $0.50 per clip, N clips charged as N \u00d7 $0.50 via Stripe quantity.",
      "tags": [
        "clipping",
        "video-editing",
        "vertical",
        "tiktok",
        "reels",
        "shorts",
        "social",
        "batch"
      ],
      "examples": [
        "Cut clips 1, 2, and 4 from this find_clips job as vertical clips",
        "Turn these three moments into TikTok-ready vertical clips in one batch",
        "Batch-cut the clips I picked from the last find_clips run"
      ]
    },
    {
      "id": "clips-horizontal",
      "name": "Clip Pack, Horizontal",
      "description": "Cut up to 5 horizontal 16:9 clips from a prior find_clips job in one batch purchase \u2014 you choose which candidates to cut, each with an optional title, delivered as separate downloadable files. One purchase, one job id, no separate extract_clip call per clip to track. Use this when you want several distinct clips; use extract_clip with multiple segments instead when you want those moments concatenated into a single highlight reel, which is one file at the single-clip price. No publishing. Cost: $0.50 per clip, N clips charged as N \u00d7 $0.50 via Stripe quantity.",
      "tags": [
        "clipping",
        "video-editing",
        "horizontal",
        "youtube",
        "batch"
      ],
      "examples": [
        "Cut clips 1, 2, and 4 from this find_clips job as separate horizontal clips",
        "Batch-cut the three moments I picked as individual files",
        "Give me each of those highlights as its own mp4, not one reel"
      ]
    }
  ],
  "protocols": [
    {
      "type": "MCP",
      "url": "https://api.weftly.ai/mcp",
      "transport": "streamable-http",
      "description": "Model Context Protocol endpoint for Claude Desktop and MCP clients"
    },
    {
      "type": "REST",
      "url": "https://api.weftly.ai/api/v1",
      "openapi": "https://api.weftly.ai/openapi.json",
      "description": "REST API with OpenAPI 3.1.0 spec"
    }
  ],
  "authentication": {
    "schemes": [
      {
        "type": "bearer",
        "description": "Anonymous session token obtained via POST /api/v1/sessions. Used after Stripe SPT payment."
      },
      {
        "type": "mpp",
        "description": "Machine Payments Protocol (MPP). Two rails: Stripe SPT (agent-initiated card payments) and Tempo (USDC crypto for fully autonomous flows). Discovery at /.well-known/mpp.json.",
        "protocol": "https://mpp.dev",
        "methods": [
          "tempo.charge",
          "stripe.charge"
        ]
      }
    ]
  },
  "pricing": {
    "model": "per-request",
    "currency": "USD",
    "items": [
      {
        "skill": "transcribe-audio",
        "price": "$0.50",
        "unit": "per file"
      },
      {
        "skill": "transcribe-video",
        "price": "$1.00",
        "unit": "per file"
      },
      {
        "skill": "summarize-audio",
        "price": "$0.75",
        "unit": "per file"
      },
      {
        "skill": "summarize-video",
        "price": "$1.25",
        "unit": "per file"
      },
      {
        "skill": "find-clips-video",
        "price": "$2.00",
        "unit": "per video"
      },
      {
        "skill": "extract-clip",
        "price": "$0.50",
        "unit": "per clip"
      },
      {
        "skill": "extract-vertical-clip",
        "price": "$0.50",
        "unit": "per clip"
      },
      {
        "skill": "publish-youtube",
        "price": "$1.75",
        "unit": "per video"
      },
      {
        "skill": "clips-vertical",
        "price": "$0.50",
        "unit": "per clip"
      },
      {
        "skill": "clips-horizontal",
        "price": "$0.50",
        "unit": "per clip"
      }
    ]
  },
  "defaultInputModes": [
    "application/octet-stream"
  ],
  "defaultOutputModes": [
    "application/x-subrip",
    "text/plain",
    "text/vtt",
    "application/json"
  ]
}