puntorigen

@puntorigen source repo

43 skills
0 followers
15 hours ago last updated

▌
Bg Music · puntorigen bundle
Generate background music tracks for social media posts, presentations, reels, and app content using ElevenLabs Music on Replicate. Supports emotional mood presets (uplifting, calm, dramatic, playful, etc.), configurable duration (up to 5 minutes), multiple output formats (MP3/WAV), and fade-in/fade-out. Use when the user asks to generate, create, or produce background music, BGM, backing tracks, mood music, or any musical audio content for posts, reels, presentations, or videos.

0
installs
▌
Broll Core · puntorigen bundle
Internal shared library for the broll-* B-roll skills (broll-web-capture, broll-browser-recorder, broll-terminal, broll-demo-avatar). Provides the canonical reel geometry/fps that match avatar-reel-composer, thin ffmpeg/ffprobe wrappers, the numbered-clip + manifest.json conventions, and the avatar picture-in-picture (PiP) compositor (pip_overlay, avatar- or base-driven). NOT invoked directly — it is imported by the other broll-* skills. Do not run this skill to make content; use broll-web-capture or broll-terminal.

0
installs
▌
Seedance 2 · puntorigen bundle
Generate video with ByteDance Seedance 2.0 via the Higgsfield MCP (generate_video, model seedance_2_0). A unified multimodal model that turns a text prompt plus reference images, video clips, and audio into audio-synced video up to 15s, with strong physics, multi-shot camera planning, and lip-sync. Use to generate video from text, animate a still image (image-to-video), do motion transfer / character consistency / style reference / audio-driven generation from references, or animate a storyboard sheet into a movie. Reframes prompts with [Image1]/[Video1]/[Audio1] tokens for references. Trigger when the user mentions seedance, seedance-2, seedance 2.0, generating a video (of a given duration/resolution/ratio), animating a storyboard, or video with image/video/audio references.

0
installs
▌
Bg Music Hq · puntorigen bundle
Generate high-quality background music and jingles using MiniMax Music 2.5 on Replicate. Supports structure tags, lyrics, 27 mood presets including pet-content moods (pet-playful, pet-heartfelt, pet-adventure, pet-chill, pet-regal, pet-goofy, pet-transformation, pet-daily, pet-epic, pet-lullaby, pet-trendy) plus presentation/podcast moods. Configurable sample rate/bitrate, WAV output for production, and fade-in/fade-out. Ideal for presentations, podcast jingles, keynote backing tracks, branded per-pet social media music, reel background tracks, and professional audio content. Use when the user asks to generate high-quality music, presentation music, podcast jingles, keynote tracks, pet-branded background music, reel music for pets, or any professional-grade musical audio content.

0
installs
▌
Broll Story · puntorigen bundle
Generate a SILENT B-roll clip of an existing avatar (and optional invented side-characters) DOING a demonstrative activity — not a talking head, no lip-sync — for an avatar reel. Given a short script, the avatar's reference image and the reel ratio (9:16 or 16:9), it authors ONE 6-panel storyboard sheet with the gpt-image-2 skill and animates that single sheet into an 8s 720p MUTED video with the seedance-2 skill. The avatar's voice is added later as voice-over by avatar-reel-composer (broll_source: existing), so these segments are intentionally silent and focus on what the avatar is DOING (POV, seen from behind, cooking, building, interacting with other characters), not on dialogue. Use when the user wants an avatar action/demonstration B-roll, a non-talking-head avatar scene, a "show don't tell" avatar sequence, or to animate the avatar doing an activity for a reel.

0
installs
▌
Cuento Reel · puntorigen bundle
Produce a "cuento" (story) reel where each avatar is a CHARACTER in a narrated tale (Narnia-style), not a single presenter talking to camera. An unseen storyteller NARRATOR (voice-only, older "cuentacuentos"/grandfatherly cloned voice) carries the story in third person, while invented character avatars ACT: silent animated action beats (broll-story / seedance image-to-video) under the narrator's voice-over, and key lines delivered as real lip-synced p-video-avatar clips (avatar-talking-video) woven in as guest scenes. Orchestrates avatar-invent, broll-story, audio-theater, avatar-talking-video and avatar-reel-composer (assemble_narration + compose_reel). Use when the user wants a story/cuento reel, to "contar una historia / cuento" with characters and a narrator, to adapt a tale (fairy tale, public-domain classic, etc.) into a narrated multi-character reel, or to make another episode in a cuento series like narnia/.

0
installs
▌
Gpt Image 2 · puntorigen bundle
Generate and edit images with OpenAI GPT Image 2 (openai/gpt-image-2) on Replicate. Excels at precise instruction-following, in-image text, photorealism, and character consistency across images using reference pictures. Use to generate images with gpt-image-2, edit/compose images from references, build character reference sheets / turnarounds from 1-3 reference pictures, build product reference sheets (4-view) from a product or product photo, or produce multi-panel storyboard sheets from a story. Defaults to high quality (the model's native maximum resolution) and 16:9. Trigger when the user mentions gpt-image-2, gpt image 2, a character sheet, character turnaround, a product sheet / product reference sheet, or a storyboard sheet, or asks to generate/edit images via Replicate.

0
installs
▌
Voice Clone · puntorigen bundle
Clone a narrator's voice from a clean voice audio file using MiniMax voice-cloning on Replicate, then generate new speech (TTS) in that cloned voice with MiniMax speech-2.8-hd. Cloning takes the clean voice MP3/WAV (e.g. voice_concat.mp3 from the voice-isolate skill), trains a voice (default speech-2.6-hd) and saves voice_id + a preview under <avatar>/voices/<name>.json. Generation reuses the avatar's trained voice (or trains one if missing), auto-detects the text language for MiniMax's language_boost, supports per-line emotion and expressive interjections ((laughs), (sighs), …) plus manual <#x#> pauses, and saves audio under <avatar>/generated-audios/ with a manifest.json. Use when the user wants to clone a voice, create/train a TTS voice, get a voice_id, or GENERATE speech / narration / audio in a cloned voice, or mentions "clonar la voz", "voice clone", "voice_id", "entrenar la voz", "generar audio", "text to speech", or "TTS con la voz".

0
installs
▌
Broll Cursor · puntorigen bundle
Turn a declarative session JSON into a short, realistic animated IDE agent-chat B-roll clip (a Cursor-style "Agent" panel) for technical reels — a user prompt typed with human jitter into the composer, then a streaming assistant turn with tool-call rows (skills) that spin then resolve with a check, in a themed high-DPI window. Renders deterministically with HTML/CSS + Playwright (frame-exact, reproducible) and NEVER runs a real agent or shell. Optionally composites the talking avatar in a PiP corner (the "base + avatar overlay" architecture). Use when you want B-roll of an AI coding agent / Cursor chat, an agent prompt + streamed skill/tool calls shown on screen, or "the request that built this" proof footage, to overlay under narration in a short video. Sibling of broll-terminal (shell) — use this one for IDE/agent-chat, that one for a command line.

0
installs
▌
Broll Finder · puntorigen bundle
Find real public YouTube footage about a topic or person (e.g. Anthony Bourdain) and turn the relevant moments into clean 9:16, silent B-roll clips for an avatar reel. The found-footage counterpart of broll-generator — instead of synthesizing B-roll with AI, it searches YouTube, fetches timecoded transcripts so the agent can pick the relevant windows WITHOUT downloading whole videos, downloads only those segments, and normalizes them into a broll/-compatible manifest (with source + license metadata) that avatar-reel-composer can drop in via a broll scene. Use when the user wants real/archival B-roll, complementary footage about a subject or person, "find clips of X on YouTube to use in the reel", or footage inserts to lay under an avatar voice-over.

0
installs
▌
Reel Restyle · puntorigen bundle
Analyze ONE avatar's reels and re-apply their style to a DIFFERENT avatar. Distills a script-agnostic reel TEMPLATE from a reference avatar's existing video-scene-analysis (scene/beat sequence, talking-head vs B-roll pattern, per-beat camera angle, cut/zoom transitions, SFX placement, caption style and proportional pacing), then applies that template to a NEW avatar given only a picture + a voice sample + a script -- auto-scaffolding the new avatar's camera-angle stills, cloned voice, talking profile and copied caption/transition styles, and drafting a composer-ready storyboard. Use when the user wants to reuse one avatar's reel style/structure (e.g. "make a reel for this new person in lolo's style", "same camera angles, transitions, SFX and timing but a different face/voice/script", cross-avatar style transfer, reel templates).

0
installs
▌
Audio Theater · puntorigen bundle
Turn a dialogue, a story with dialogue, or a one-line idea into multi-character audio using Gemini TTS, with word-level timecoded transcription (WhisperX or local faster-whisper), realistic ambient/foley SFX (ElevenLabs or the sound-effects skill), and instrumental score/music (MiniMax Music 2.6). Optional stereo spatialization (pan + distance + movement) and an ffmpeg mix with sidechain ducking (music ducks under voices+SFX). Three modes - theater (dramatized radio play), lipsync (clean per-line clips <=15s plus a manifest for the seedance-2 skill), and podcast (two hosts with intro/bed/outro music). With music it delivers three tracks (full mix, music-only, and a no-music stem for animated storyboards) plus a timecoded transcript and per-line voice + SFX/music files. Use for an audio drama, audio theater / audioteatro / radioteatro, a dramatized dialogue, a score for a story/podcast/storyboard, voiceover for lipsync, a TTS conversation, or a podcast episode from a script or idea.

0
installs
▌
Avatar Frames · puntorigen bundle
Extract ready avatar reference frames from talking-head videos — single face, face-region sharpness, no burned-in subtitles. Optionally saves inpaint candidates (1 face, sharp, with subtitles) in with_subtitles/ when requested or when no ready frames exist. Uses MediaPipe and EasyOCR. Use when the user asks to extract avatar frames, clean frames from video, reference images for virtual avatar, or talking-head frame extraction.

0
installs
▌
Avatar Invent · puntorigen bundle
Invent a brand-new FICTIONAL avatar from a text description — no real person, photo or recording needed. Generates a front-facing, well-lit, seated half-body presenter still (photorealistic by default; soft3d/Pixar, anime or a custom style on request) in a room that fits the topic, derives camera-angle variations, and DESIGNS a matching voice with ElevenLabs Voice Design which is then cloned with MiniMax — all written into the same per-avatar folder structure (scene.json, talking_profile.json, refs/, angles/, voices/, avatar.json) that avatar-reel-composer and reel-restyle consume. Defaults follow UGC talking-head reel best practices (eyes to lens, soft key from the left + soft backlight, vertical 9:16). Use when the user wants to make up / invent / fabricate an avatar, character or presenter for a reel from just a description (e.g. "inventa un avatar de una psicóloga para un reel", "create a fictional presenter", "necesito un personaje para generar un reel"), rather than cloning an existing person.

0
installs
▌
Sound Effects · puntorigen bundle
Generate consistent sound effects and short audio assets for mobile/web apps using Stable Audio 2.5 on Replicate. Supports UI sounds, notifications, game SFX, ambient loops, transitions, and more. Automatically enhances prompts for professional-quality results. Use when the user asks to generate, create, or produce sound effects, audio assets, UI sounds, notification tones, or any short audio content for their app.

0
installs
▌
Tiktok Videos · puntorigen bundle
Download all public videos from a TikTok profile (or a single TikTok video / short link) to a local folder, named with upload date-time (YYYY-MM-DD_HH-MM-SS.mp4). Uses yt-dlp as the primary engine with the tikwm.com no-watermark API as a fallback. Use when the user asks to download TikTok videos, save a TikTok profile's videos locally, grab a TikTok without watermark, or batch-download profile videos with datetime filenames.

0
installs
▌
Video Compose · puntorigen bundle
Compose a polished video reel from a folder of source clips and images using a Python+FFmpeg pipeline with Remotion title overlays. The skill drafts an interactive treatment (shot list) with the user, analyzes available assets via ffprobe + PySceneDetect + Gemini Vision, generates branded background music via bg-music-hq, builds a beat-synced Edit Decision List (EDL), renders a low-res preview for approval, then renders animated titles via Remotion (ProRes 4444 with alpha) and the final composite via FFmpeg with Ken Burns motion, xfade transitions, and music mixing. Use when the user asks to compose, edit, assemble, or stitch a video reel from a folder of clips and photos, especially when they want titles, transitions, background music, or VEED.io / Adobe-Premiere-style auto-editing without a voiceover.

0
installs
▌
Voice Isolate · puntorigen bundle
Extracts clean voice samples from a video. Separates the vocal stem with Demucs, gets speech timecodes with faster-whisper, detects the recurring percussive "taka" SFX by its high-frequency signature in the accompaniment stem, drops the whole word under each detected SFX, and exports the remaining clean spoken phrases as individual sample clips plus a concatenated track. Use when the user gives a video and wants the narrator's voice isolated, clean voice samples for cloning/TTS, a voice-only audio track, background music/SFX removed from speech, or mentions "solo la voz", "aislar la voz", "voz limpia", "muestras de voz", "quitar SFX/música", "canal de voz", isolate/extract voice, voice samples, narrator voice, or vocal stem from a video or reel.

0
installs
▌
Broll Terminal · puntorigen bundle
Turn a declarative session JSON into a short, realistic animated-terminal B-roll clip for technical reels — commands typed with human jitter (variable per-key delay, the odd typo+backspace), output that streams in with spinners and ANSI colors, a macOS/Warp-style window, themed and high-DPI. Renders deterministically with HTML/CSS + Playwright (frame-exact, reproducible) and NEVER executes a real shell. Optionally composites the talking avatar in a PiP corner (the "base + avatar overlay" architecture). Use when you want B-roll of terminal commands, a CLI demo, code/agent commands shown on screen, or animated proof footage of a tool/pipeline running, to overlay under narration in a short video.

0
installs
▌
Reel Discovery · puntorigen bundle
Find highly-ranking PUBLIC short videos/reels by TOPIC (keyword/hashtag) or by BUSINESS (brand name/handle) across YouTube, TikTok, Instagram and Facebook, then rank them by views/engagement/velocity into one unified manifest. Free-first (YouTube Data API + tikwm + Instagram web endpoints) with an optional paid Apify upgrade for robust TikTok keyword search, Instagram topic search and all Facebook discovery (Facebook has no free anonymous path). Optionally downloads the top N MP4s ready for the video-scene-analysis -> reel-restyle pipeline. Use when the user wants to discover trending/top/viral reels, research a competitor's or topic's best-performing short videos, find reference reels to analyze or restyle, or "what reels are working for <topic/brand> on TikTok/YouTube/Instagram/Facebook".

0
installs
▌
Youtube Videos · puntorigen bundle
Download all public videos from a YouTube channel (or a playlist, single video, or Short) to a local folder, named with upload date-time (YYYY-MM-DD_HH-MM-SS.mp4), capped at 720p mp4. Uses yt-dlp + ffmpeg. Use when the user asks to download YouTube videos, a channel's videos, a playlist, Shorts, or batch-download videos with datetime filenames.

0
installs
▌
Asset Generator · puntorigen bundle
Generate image assets for web and mobile apps using Google Gemini 3 Pro Image (Nano Banana Pro). Supports style presets (icon, illustration, logo, mascot, hero, marketing, infographic, etc.), transparent backgrounds via ML removal, multiple formats (PNG, WebP, JPEG), up to 14 reference images, native 4K resolution, and platform-specific export (iOS icons, Android icons, favicons, PWA, social media). Use when the user asks to generate, create, or produce images, icons, illustrations, logos, mascots, app assets, or any visual content for their app or website.

0
installs
▌
Avatar Location · puntorigen bundle
Create a new LOCATION (a look) for an existing avatar — keep its IDENTITY (face, gestures, cloned voice, talking_profile) but vary the LOOK: wardrobe + environment + light, optionally incorporating asset references (a logo on a shirt/jacket, a prop in the scene). Generates an identity-anchored hero still + camera angles under <avatar>/locations/<loc>/, selectable later per reel and per scene by avatar-reel-composer, so one avatar can appear in many looks without re-making it. The avatar's DEFAULT location is just what exists today (top-level scene.json + angles/) and is untouched. Use when the user wants to "dress" an avatar differently, put it in another setting/room, give it a wardrobe/scene variant, add a branded outfit (logo on the shirt) or a prop, create reel looks, or asks for avatar locations / outfits / wardrobes / environments / "otro ambiente" / "otra ropa" for an avatar.

0
installs
▌
Broll Generator · puntorigen bundle
Generate hyper-realistic complementary B-roll video clips (3-6s) from a scene description using Pruna p-video on Replicate, with NO main character — only people, objects and situations that reinforce an idea. Clips are vertical 9:16 at 720p by default and silent (clean for voice-over), so an avatar's narration can be laid on top in a composite reel. Use when the user wants to create B-roll, inserts, cutaways, complementary footage, or realistic supporting clips for a talking-head/avatar video or reel.

0
installs
▌
Facebook Videos · puntorigen bundle
Download public Facebook videos (a watch link, video permalink, reel, fb.watch short link, or a Page's videos tab) to a local folder, named by upload datetime (YYYY-MM-DD_HH-MM-SS.mp4). Single-engine yt-dlp + ffmpeg; uses --cookies-from-browser for login-walled public videos. Use when the user asks to download a Facebook video or reel, save an FB video locally, grab a video from a facebook.com/watch or fb.watch link, or batch-download a Page's videos.

0
installs
▌
Broll Actor Copy · puntorigen bundle
Generate a B-roll clip where OUR avatar COPIES the motion of a driving video — same movement, gestures, expressions and lip movements — via ByteDance DreamActor M2.0 (bytedance/dreamactor-m2.0 on Replicate). Takes an avatar, a location/look (or its default) and a driving video, resolves the look's hero image, and transfers the video's performance onto it. Works with humans, cartoons and animals. The driving video must show exactly ONE animated character; multi-character sources are split into single-character segments and stitched. A human driver should match a human avatar's gender (else warn + ask: the face distorts) and the reference scene should include the objects the driver interacts with. Output keeps the reference image's resolution (9:16 hero → 9:16 clip); the clip is muted and avatar-reel-composer re-lays narration. Use when the user wants the avatar to mimic a video's movement, a motion copy / performance transfer / actor copy / "copiar el movimiento del video" B-roll, or mentions dreamactor.

0
installs
▌
Instagram Videos · puntorigen bundle
Download all public videos/reels from an Instagram profile to a local folder, named with upload date-time (YYYY-MM-DD_HH-MM-SS.mp4). Uses Picnob via browser automation plus a Python download script. Use when the user asks to download Instagram videos, reels, or video content from a public profile, save IG videos locally, or batch-download profile reels with datetime filenames.

0
installs
▌
Rule Of Six Edit · puntorigen bundle
Rule of Six — reel edit guide

0
installs
▌
Video Bg Replace · puntorigen bundle
Replace the background of a talking-head clip by matting the speaker (Robust Video Matting on Replicate) and compositing them over a NEW background — a generated/animated B-roll clip, a provided video, or a still image. The speaker layer gets a temporally-stable alpha matte, then ffmpeg lays it over the new background (alpha-mask + original RGB, or green-screen + chromakey), with optional edge feather, grounding drop-shadow and a unifying color grade, while preserving the speaker's original audio. Use when the user wants to swap/replace a video background, put an avatar over a different scene, place an animated video behind a person, key out / matte a talking head, or composite a presenter onto generated B-roll.

0
installs
▌
Video Transcribe · puntorigen bundle
Download a single video by URL (Instagram reel/post, or any yt-dlp-supported link) and transcribe its spoken audio to text with faster-whisper — or transcribe a local media file. Writes plain text, an SRT subtitle file, and a JSON with per-segment timecodes. Uses yt-dlp for download and faster-whisper (default `small`, CPU/int8) for transcription. Use when the user wants the spoken text / transcript / subtitles of an Instagram reel or other video, wants to turn a reel's audio into a script, or wants to transcribe a local audio/video file.

0
installs
▌
Broll Web Capture · puntorigen bundle
Turn a website or a GitHub repo into a short, polished B-roll clip for technical reels — a high-DPI Playwright capture animated with Ken Burns zoom/pan, a vertical scroll-reveal, or a spotlight push-in, framed on the target aspect (9:16 etc.) with a blurred-fill background. A github preset captures repo money-shots (landing/README/contribution graph) with a live animated star counter pulled from the GitHub API. Optionally composites the talking avatar in a PiP corner (the "base + avatar overlay" architecture). Use when you want B-roll showing a website, landing page, product page, or GitHub repository, a screen capture with zoom-in of a repo/site, or animated proof footage to overlay under narration in a short video.

0
installs
▌
Onboarding Script · puntorigen bundle
Onboarding Script

0
installs
▌
Avatar Ambient Sfx · puntorigen bundle
Add a spatial ambient SFX layer to a finished avatar reel and deliver it as final-espacial.mp4, without overwriting final.mp4. Generates positioned ambient/foley beds + one-shots (birds, water, gulls, fire crackle, crickets, wind, etc.) with audio-theater (ElevenLabs Sound Effects), mixes them on a virtual stereo stage (pan + distance + movement) with mix_spatial.py, and overlays them ducked under the narration. Use when the user wants to "ambientar más" / add ambient or positional/spatial sound to a reel, or asks for sound design / atmosphere that follows different worlds/locations in a video. Builds on the audio-theater skill.

0
installs
▌
Carousel Generator · puntorigen bundle
Research the winning carousel structure of Instagram competitors and generate on-brand carousels (per-slide images + caption + hashtags) for any brand. Scrapes the real carousels of a Top-10 competitor list via Apify, proves whether carousels beat reels, derives a slide-by-slide blueprint with Gemini vision, fuses it with reel-discovery narrative patterns, then produces 4:5 slides with Gemini 3 Pro Image (asset-generator), keeping brand palette/typography/identity consistent across slides. Use when the user asks to create an Instagram carousel, study what carousels a competitor/niche is winning with, replicate a winning carousel structure, or produce branded carousel slides with copy and hashtags.

0
installs
▌
Viral Video Script · puntorigen bundle
Write short-form video scripts / dialogue that follow Tuan Le's "3 billion views" viral formula — the six psychological principles (familiarity / format-steal, curiosity-gap hook, identity over product via the Means-End Chain, the credential / authority shortcut, built-in shareability, and the Hook→Problem→Story→Payoff story skeleton with fast cuts + captions). Produces a beat sheet (JSON), a shooting script (Markdown) and a clean narration track (txt) ready to feed voice-clone / avatar-reel-composer. Use when the user wants to write a script, dialogue, hook, or voice-over for a reel / short / TikTok, make a brand or product go viral, repurpose a boring product into engaging content, or mentions "guion", "guión", "script para un reel", "diálogo", "hook viral", "voz en off", or the viral / psychology content formula.

0
installs
▌
Broll Avatar Camera · puntorigen bundle
Generate a short B-roll clip of OUR OWN avatar DOING something on camera — an ACTION shot that COMPLEMENTS the narration (movement matters more than words), not a centered talking head. Built on the SAME model as our talking-heads (prunaai/p-video-avatar) so the avatar's face/wardrobe/room stay consistent. Pipeline: build ONE action start frame with the gpt-image-2 skill (locking wardrobe/room/light from the scene profile), then animate it with p-video-avatar driven by the beat's narration audio, with the action in video_prompt and exclusions in negative_prompt; the clip is muted and the master narration is re-laid by avatar-reel-composer. For a face-free beat or a precise object move with exact start+end poses, use seedance-2 instead. Use when the user wants the avatar SEEN doing an activity in-scene (reaching, tending, building, holding an object) under a narration line, or mentions broll-avatar-camera, p-video-avatar action shots, video_prompt/negative_prompt, or an action insert of the real avatar.

0
installs
▌
Caption Word Reveal · puntorigen
Word-by-word ("karaoke") burned-in captions for avatar reels: each word appears at the instant it's spoken, building up the SAME phrase unit in place (frozen layout, no reflow), and the phrase clears when the next one begins. Use when the user wants captions/subtitles where words form the sentence as the narration says them (word-by-word reveal, karaoke captions, "las palabras van apareciendo") on any reel produced by the avatar-reel-composer pipeline.

0
installs
▌
Avatar Camera Angles · puntorigen bundle
Generate realistic camera-angle variations of a talking-head avatar from a single reference frame, reusing the gpt-image-2 skill. Produces the SAME person, outfit, room and lighting seen from a different virtual camera position (push-in, pull-out, low/high angle, three-quarter, Dutch tilt, off-center negative space for captions) so a reel can "cut" the camera every 4-8s while the speaker keeps addressing the lens. Each still then drives a lip-synced clip (seedance-2 / VEED Fabric). Use when the user wants to create reel shots / camera cuts / multiple angles / perspectives for an avatar or talking-head video, simulate a multi-cam talking head, or vary the framing of a presenter while keeping identity and scene consistent.

0
installs
▌
Avatar Reel Composer · puntorigen bundle
Turn a script + an existing avatar into a finished reel — 9:16 vertical (TikTok/Reels) or 16:9 landscape (YouTube), set via the storyboard `format`. Narrates the script once in the avatar's cloned voice, cuts it per scene by word-level alignment, generates lip-synced talking-head scenes (avatar-talking-video) and silent B-roll scenes (broll-generator) as voice-over, applies Ken Burns / zoom motion, and assembles with hard cuts under one master narration. Can weave in a GUEST/cameo scene (a clip of a DIFFERENT avatar speaking in its own voice) that keeps its own audio. Supports per-reel and per-scene LOCATIONS (alternate avatar looks built by avatar-location). Can also onboard a brand-new avatar from a public Instagram profile via create_avatar.py (download, analysis, frames, voice cloning, style profiling). Use when the user wants to produce a full reel/short from a script for an avatar that already exists (videos/, voices/, angles/ and/or talking_profile.json), or to onboard a new avatar from an Instagram UR

0
installs
▌
Avatar Talking Video · puntorigen bundle
Generate a talking-head avatar video from a line of text. Stage 1 synthesizes the text in the avatar's cloned voice via the voice-clone skill (reusing the avatar's trained voice, or training one from a source recording if it has none). Stage 2 feeds a camera-angle image (from avatar-camera-angles) plus the generated mp3 to prunaai/p-video-avatar on Replicate at 720p (or 1080p) and saves the lip-synced MP4 under <avatar>/generated-videos/ with a manifest.json. Use when the user wants a talking avatar / talking-head video, a lip-synced clip of an avatar saying some text, to animate a camera-angle still into a speaking video, or mentions "video del avatar hablando", "avatar que hable", "talking head", "lip sync", or p-video-avatar.

0
installs
▌
Video Scene Analysis · puntorigen bundle
Analyze local video files (mp4, mov, webm) into scene sequences: scene-change detection, scene type (talking head vs B-roll), split-screen / screen-composition detection (B-roll band + presenter, picture-in-picture, graphic overlays), B-roll kind (archival footage of a recognizable/known person vs generic material, naming the people), presenter background (real set/location vs animated/cartoons/motion graphics), zoom in/out vs previous scene, faster-whisper transcription with timecodes, SFX/music-bed detection per scene, a representative frame per scene, and per-scene focus/emotion summaries. Outputs .analysis.json, .analysis.md, and a frames folder. Use when the user asks to analyze a video, detect scenes, camera angle, split-screen layouts, archival/known-person footage, animated vs real backgrounds, transcribe a local video, detect SFX or background music, build a shot list from footage, or understand reel/avatar video structure.

0
installs
▌
Brand Content Strategy · puntorigen bundle
Turn a website or personal brand into a research-backed content + channel strategy. Profiles the site's positioning/credentials, discovers what reels rank in its niche (reel-discovery), filters out official/brand accounts to rank only real individual creators, reverse-engineers their cross-platform distribution playbook (perplexity_ask), and writes a positioning + 30-day content plan (with an optional AI-voice podcast layer wired to voice-clone + audio-theater). Use when the user wants to promote/grow a website, get known, position themselves as an expert/authority, decide what videos to make, launch a YouTube/TikTok/Instagram channel, find a target audience, or build a content/creator strategy for a domain or brand.

0
installs
▌
Broll Browser Recorder · puntorigen bundle
Record a real browser product demo as smooth continuous video (Playwright HTTP server, adaptive agent-driven interaction), post-process it for a reel in 9:16 vertical or 16:9 landscape/YouTube (auto-camera reframe that follows the interactions, trim/speed/preset), and optionally composite a talking avatar in a picture-in-picture corner that NARRATES what the demo shows. Unlike broll-web-capture (a Ken Burns still), this is a live click-through demo — the recorded demo is the hero and drives the clip length (base-driven PiP); the avatar is authored afterwards to talk about it. Use when you want a real product/app walkthrough as the base layer of a reel, a browser demo with a presenter narrating in the corner, or a screen-recorded click-through reframed to 9:16 or 16:9.

0
installs

43 published skills