mordor-forge/gemini-media

Generates and edits images with Google's Nano Banana models through the gemini-media MCP server - text-to-image, photo edits and retouching, multi-reference composition, product shots, posters, logos and infographics with legible text, stickers and icons, variations, and search-grounded visuals of current events or real places. Use when the user wants to create, draw, design, mock up, restyle, relight, extend or combine pictures, or needs keyframes for a video, even if they never mention Gemini. Not for charts plotted from data, SVG or code-drawn graphics, video clips (gemini-video) or audio.

Plans and produces multi-asset media projects with the gemini-media MCP server by chaining Nano Banana images, Veo or Gemini Omni video, Gemini TTS voiceover and Lyria music into one deliverable - product promos, social ads, short story or explainer videos, trailers, podcast intros. Covers shot lists, consistent characters across shots, an up-front budget with cost estimates and approval, parallel video jobs, and assembly with ffmpeg. Use when a request needs two or more media types or several shots combined, even if the user just says "make a promo video". For a single image, clip, voiceover or track, use gemini-image, gemini-video, gemini-speech or gemini-music instead.

Composes original music with Google Lyria through the gemini-media MCP server - 30-second clips, loops, jingles and stingers, or full songs of a few minutes with vocals, custom lyrics in any language, section structure and timing control, instrumental beds for videos and podcasts, and music inspired by images. Use when the user wants a song, beat, soundtrack, theme tune, background or intro music, or describes a musical idea to hear, even without naming Lyria. Not for spoken voiceover or narration (use gemini-speech), sound effects inside a video clip, or editing, remixing or extending an existing audio file.

Generates natural speech audio (WAV) with Gemini text-to-speech through the gemini-media MCP server - voiceovers, narration, audiobook passages, announcements, explainer and ad reads, and two-speaker podcast or dialogue scripts, with controllable delivery style, 30 prebuilt voices, inline sounds like sighs and pauses, and automatic language detection. Use when the user wants text read aloud, a voiceover or narration track, TTS, a spoken intro, or a conversation rendered as audio. Not for music or singing (use gemini-music), transcription or speech-to-text, dialogue inside a generated video clip (gemini-video), or live voice chat.

Generates and edits short video clips with native audio using Google Veo and Gemini Omni Flash through the gemini-media MCP server - text-to-video, animating a still image, first-to-last-frame transitions, keeping a character or product consistent from reference images, changing an existing clip with an instruction, and extending clips into longer shots. Handles the asynchronous job flow, resolution and duration rules, and cost approval for this expensive medium. Use when the user wants a video, clip, animation, b-roll, cinematic shot or product spin, wants to bring a picture to life, or wants to add, remove or restyle something in a short clip. Not for trimming or cutting footage, screen recordings, slideshows, or audio-only work.