marswaveai/skills▌

Generate podcast episodes with 1-2 AI speakers discussing a topic. Supports quick overviews, deep analysis, and debate formats. Input can be a topic description, URL(s), or text. Output is a full audio episode with transcript.

Extract and normalize content from URLs across supported platforms. Returns structured data including content body, metadata, and references. Useful as a preprocessing step for content generation skills or standalone content extraction.

Generate AI images using the Labnana API. Supports text prompts with optional reference images, multiple resolutions, and aspect ratios. Images are saved as local files.

Convert text to natural-sounding speech with single or multi-speaker audio generation. \n \n Two modes: Quick mode for instant single-voice MP3 output, and Script mode for multi-speaker dialogue with per-character voice assignment \n Automatic mode detection based on input structure; supports both plain text and structured scripts with character markers \n Built-in speaker selection with language support (Chinese and English) and preference saving to local config \n Configurable output modes: in

Deprecated skill — migrate to individual focused skills for podcasts, videos, TTS, and image generation. \n \n This skill has been replaced by separate, specialized skills: /podcast , /explainer , /tts , /image-gen , and /content-parser \n Users should upgrade by running npx skills add marswaveai/skills and restarting their agent \n Do not execute requests until the user has completed the migration to the new skill set \n

Generate explainer videos that combine a single narrator's voiceover with AI-generated visuals. Ideal for product introductions, concept explanations, and tutorials. Supports text-only script generation or full text + video output.

Local offline audio transcription with multi-language support and optional AI polishing. \n \n Transcribes audio files to text using coli asr with no API keys required; supports Chinese, English, Japanese, Korean, and Cantonese via sensevoice model, or English-only via whisper-tiny \n Models download automatically on first use (~60MB) to ~/.coli/models/ ; requires coli CLI and ffmpeg (WAV files work without it) \n Optional AI polishing step corrects punctuation, removes filler words, and improve