vieiraae/imagegen-sidekick
Skills for generating and editing images (including Blender-ready assets), synthesizing speech with SSML and word timelines, and recording narrated web UI walkthrough videos.
Create SSML and synthesize speech with Microsoft Foundry Speech (MAI-Voice, default Harper), producing the SSML, an audio file and a JSON word timeline. Use for narration, voice-overs, dialogue, audiobook or podcast clips, lip-sync and subtitle timing, and Blender/video audio tracks — even when the user just says "read this aloud", "make a voice line" or "I need TTS".
Generate or edit images with GPT Image on Azure OpenAI. Use for requested illustrations, photos, icons, transparent cutouts, reference-image compositions, masked edits, and Blender textures, decals, billboards or modeling references. Not for creating 3D meshes or merely discussing image APIs.
Record narrated walkthrough videos of web apps. Narration comes from foundry-tts; Playwright browser actions follow the spoken word timeline, with highlights, labels, an animated cursor, optional gpt-imagegen images and karaoke captions, producing an MP4. Use for product demos, UI tutorials, onboarding or how-to videos, feature tours and screen-recorded explainers of a website or local web app, even if the user just says "make a video showing how to use this page".