{"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.dev.weftly.ai","version":"1.0.0","provider":{"organization":"Woven Record Media","url":"https://weftly-landing.pages.dev"},"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, edit (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, edit (per-word timestamps, schema at /.well-known/weftly-transcript-v2.schema.json) — 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, words, or edit 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, words, or edit 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 2–5 minute moments with titles, hooks, narrative summaries, scoring, and the FULL TEXT of each candidate so callers can preview content before paying for clips_from_job. 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 — 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 — 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. 1080×1920 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 — 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) — 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 — 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 (1080×1920, TikTok/Reels/Shorts-ready) from a prior find_clips job in one batch purchase — you choose which candidates to cut, each with an optional title, delivered as downloadable files. One purchase, one job id — no separate clips_from_job call per clip to track. No publishing in v1. Cost: $0.50 per clip, N clips charged as N × $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 — you choose which candidates to cut, each with an optional title, delivered as separate downloadable files. One purchase, one job id, no separate clips_from_job call per clip to track. Use this when you want several distinct clips; use clips_from_job with multiple segments and output \"reel\" 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 × $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"]},{"id":"clips-from-video","name":"Clips From Video","description":"Upload a video and get 9:16 vertical clips back in one call — finds the best moments AND cuts up to 5 of them, under one payment. Chat-native alternative to calling find_clips then clips_from_job yourself. A Stripe card hold authorizes the worst case (up to $4.50); the customer is charged once, at the end, for exactly what delivered. Card rail only — no MPP/Tempo, because a chain authorizes before the final price is known. Wallet-only callers should use find_clips then clips_from_job instead.","tags":["clipping","video-editing","vertical","tiktok","reels","shorts","social","chain"],"examples":["Turn this interview into TikToks","Find the best moments in this video and cut them as Reels"]},{"id":"clips-youtube","name":"Clips For YouTube","description":"Upload a video and get 16:9 clips back in one call — finds the best moments AND cuts up to 5 of them, under one payment, in the source's native framing. It cuts and hands back files; it does NOT publish to YouTube (publish_to_youtube is the separate product for that). A Stripe card hold authorizes the worst case (up to $4.50); the customer is charged once, at the end, for exactly what delivered. Card rail only — no MPP/Tempo, because a chain authorizes before the final price is known. Wallet-only callers should use find_clips then clips_from_job instead.","tags":["clipping","video-editing","horizontal","youtube","chain"],"examples":["Cut the best moments from this talk as YouTube clips","Find highlights in this recording and cut them as 16:9 files"]},{"id":"edit-from-transcript","name":"Edit From Transcript","description":"Cut and re-assemble a video from an edited copy of its word-level transcript. Get the editable text via get_job_status(format: \"edit\") or the download route — one [NNNN]-marked line per sentence. Delete words or whole lines, move whole lines, but do not add or change words. Send the full edited text inline (max 256 KB). Validated before payment: an invalid edit returns the failing lines and costs nothing. Up to 300 cuts and 30 minutes of output. Returns the edited video plus a re-timed SRT and word-level transcript. Cost: $1.00, paid via MPP — accepts Tempo USDC and Stripe SPT.","tags":["editing","video-editing","transcript","clipping","compose"],"examples":["Delete the filler lines from this interview transcript and cut the video to match","Move the closing answer to the top of this talk and give me the edited video","Tighten this recording by cutting the two tangents I marked for deletion"]}],"protocols":[{"type":"MCP","url":"https://api.dev.weftly.ai/mcp","transport":"streamable-http","description":"Model Context Protocol endpoint for Claude Desktop and MCP clients"},{"type":"REST","url":"https://api.dev.weftly.ai/api/v1","openapi":"https://api.dev.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"},{"skill":"clips-from-video","price":"$4.50","unit":"per video"},{"skill":"clips-youtube","price":"$4.50","unit":"per video"},{"skill":"edit-from-transcript","price":"$1.00","unit":"per edit"}]},"defaultInputModes":["application/octet-stream"],"defaultOutputModes":["application/x-subrip","text/plain","text/vtt","application/json"]}