Image & Video

Takes an existing image ad and rebuilds it at every ratio you need — recomposed, not cropped.

Composites your source video with shot-analysis data into a watchable video where the current shot's metadata panel switches in sync with every cut.

Generates WeChat Official Account cover images and in-article illustrations by matching the article's title and content to reusable style presets.

Asks 3-4 targeted questions about your video needs, then recommends the best 1-3 of 16 AI video models with clear reasoning.

Generates Kling 3.0 (Kuaishou) video prompts by auto-selecting one of three writing formulas based on your request.

Converts an AI video prompt from one model's writing style into the target model's best-practice format (Sora, Kling, Veo, Wan, Hailuo and more).

Takes a video plus an SRT and either burns the subtitles into the pixels with libass or soft-muxes a togglable track, and can mix a dub over the original audio in a single ffmpeg pass.

Breaks a finished video into measured cut rhythm, loudness and section structure, then exports a production_reference.json containing only reusable methods — no original names or dialogue.

Executes an agent-authored clip plan: validates clip_plan.json, snaps cut boundaries safely, and concatenates the selected ranges into edited_source.mp4.

Final-stage assembly skill that lays narration audio over the source video, ducks the original sound, generates and burns subtitles, and normalizes loudness into a finished recap.

Writes copy-ready image prompts with the right model, quality level, and aspect ratio for your request.

Builds 2–5 minute, 16:9, voiceover-free knowledge films where one line of subtitle is literally acted out on screen, rendered frame-by-frame with Canvas + Playwright + ffmpeg.

Turns an already-analyzed video into a director-level story plan, A/V board, and timestamped narration JSON — then validates it before TTS.

Turns a timestamped narration.json into per-segment TTS audio that fits each time window, plus placement metadata for downstream mixing.

Art-direction skill for achieving clean anime cel-shaded looks in Blender.

Produce and validate game-ready Blender animation — walk/run/idle cycles, combat, mechanical and camera motion — with MCP-driven checks.

A pre-export validation gate for Blender assets: polycount budgets, topology cleanup, UV efficiency, material/naming conventions and LOD generation.

Writes and revises copy-ready image prompts for short-drama characters, looks, locations, props and state variants in a standardized Markdown sheet.

Turns one knowledge point into a narrated 45-second-to-5-minute short film with sourced facts, styled visuals, BGM, and OCR quality control.

Turns a single topic into a full lesson-video series — research, narration, TTS, scene choreography, 1080p rendering and publishing — through one scripted pipeline.