content-machine

Plan, build, and fix short-form videos like an editor, using hook-first structure, caption-aware visual design, Remotion/FFmpeg assembly, and reject-regenerate review loops instead of one-shot automation.

Build a full short from brief to reviewed MP4 while keeping script, audio, visuals, captions, and validation aligned as one edit.

Generate and evaluate multiple short-form hook, title, cover, opening-frame, and platform-packaging variants before scripting or rendering, then choose a default and preserve the rationale.

Turn publish receipts and performance metrics into reusable style-profile or niche-profile updates that say what to repeat, kill, or test next.

Generate platform-specific publish metadata, fit checks, and upload checklists for short-form and social video without auto-uploading.

Review scripts, timelines, scene plans, or rendered shorts for retention and mobile pacing before final render approval or publish readiness.

Review a short before or after production using quality gates for hook, retention, clarity, payoff, risk, and platform fit without promising guaranteed virality.

Extract approved longform highlight candidates into clip-local video, audio, timestamps, visuals, and clip-plan artifacts after selection, boundary snap, and approval.

Turn a podcast, interview, talk, screen recording, or long YouTube video into strong vertical shorts by selecting moments, snapping cut boundaries, reframing to 9:16, and rendering with native captions.

Design and implement burned-in captions for short-form video across TikTok, Reels, Shorts, CapCut, Hormozi, karaoke, and educational styles using this repo's visual system and timing rules.

Run the repo’s environment and dependency diagnostics as a JSON-stdio skill so Claude Code or Codex can capture a health report instead of scraping CLI stderr.

Save reusable local short-form style profiles for captions, pacing, visual rules, and render defaults.

Make a faceless explainer short using diagrams, motion cards, generated scenes, typography, and light footage support instead of relying on talking heads or pure stock footage.

Turn a topic, packaging hint, or blueprint into a script file that the rest of the pack can use.

Generate multi-scene AI video sequences with better visual continuity by chaining each scene from the last frame of the previous clip and resuming cleanly when some scenes already exist.

Build a faceless short that mixes voiceover, stock footage, local footage, gameplay, AI-generated images or clips, captions, and music into one coherent edit.

Turn a short's opening hook into its own styled visual asset or overlay card so the first beat lands as designed typography, not as an afterthought subtitle.

Maintain a local JSON index of reusable source media, transcripts, tags, and probe metadata.

Build a narrow educational short around motion cards, timed reveals, quiz beats, or vocabulary-style prompts where the primary visual system is designed cards rather than stock footage or freeform generated scenes.

Build Reddit-style post cards and similar social-thread overlays as reusable visual assets so story shorts can show the post, author, votes, and comments cleanly instead of faking everything directly in the timeline.

Analyze a reference short from a local file or URL into reusable breakdown files so an agent can study what made it work.

Generate a voiceover WAV plus timestamps from a script file so later steps can build captions and video from it.

Route and quality-gate TikTok, Instagram Reels, and YouTube Shorts production using the local 2026-04-29 archetype research pack, blueprints, repo evidence summaries, workflow guides, and provenance rules.

Convert a scene plan into stronger image or video generation prompts by using explicit shot language, movement, lighting, depth, and subject texture instead of generic style prefixes.

Audit user-supplied video, audio, image, and gameplay inputs before planning or render so the agent knows what is actually in the media, what quality risks exist, and whether the input is fit for reuse.

Turn a topic, fact, or news hook into a faceless stock-footage explainer short with hard scene structure, timed visual intent, captions, music, and packaging.

Assemble the final short from visuals, timestamps, audio, and caption treatment, using the repo's Remotion stack without losing short-form visual intent.

Use when brainstorming, selecting, or verifying external creative websites for content-machine videos, including animation/component libraries, design inspiration, stock media, 3D models/textures, AI image/video tools, music/SFX, fonts, icons, and other source assets.

Use when generating or improving SVG, HTML, CSS, React, or Remotion animations for content-machine videos so motion is frame-driven, smooth, readable, performant, and safe for captions/platform chrome.

Use when creating, selecting, or adapting caption-clean 3D or procedural gameplay-like background loops for shorts without replacing the chosen content lane.

Score the script, validate the rendered video, and produce publish metadata so the agent can decide whether a short is ready to upload.

Build or update a machine-readable asset ledger for generated, local, stock, audio, gameplay, and manually assembled demo assets so publish-prep can review provenance.

Probe source media locally with ffprobe/ffmpeg and produce duration, orientation, stream, scene-change, silence, and audio-energy metadata.

Interactive longform-to-shortform video creator. Extracts viral-ready short clips from long videos using Claude as the orchestrator. Transcribes with faster-whisper (GPU), Claude scores and presents candidate segments interactively, user picks and adjusts, Remotion renders premium animated captions (Bold/Bounce/Clean styles), FFmpeg exports platform-optimized files (YouTube Shorts, TikTok, Instagram Reels). Use when user says "shorts", "short clips", "shortform", "extract clips", "tiktok from video", "reels from video", "vertical clips", or "create shorts".

Make a faceless explainer short using diagrams, motion cards, generated scenes, typography, and light footage support instead of relying on talking heads or pure stock footage.

Track generated media assets with stable file fingerprints so caches, previews, and downstream consumers can tell when storyboards, videos, thumbnails, or design assets actually changed.

Snap selected highlight candidate boundaries to nearby sentence or pause boundaries before clipping.

Turn a topic, packaging hint, or blueprint into a script file that the rest of the pack can use.

Generate multi-scene AI video sequences with better visual continuity by chaining each scene from the last frame of the previous clip and resuming cleanly when some scenes already exist.

Run the repo’s environment and dependency diagnostics as a JSON-stdio skill so Claude Code or Codex can capture a health report instead of scraping CLI stderr.