A powerful integration for Google Gemini that enables end-to-end multimodal workflows including image generation, video synthesis, and speech-to-text capabilities.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install google-gemini-media
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install google-gemini-media using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Gemini Multimodal Media (Image/Video/Speech)?
The Gemini Multimodal Media skill is a developer-centric toolkit designed to bridge the gap between text-based AI and rich media generation. By consolidating six distinct Gemini API capabilities—including Nano Banana for image creation and Veo 3.1 for high-fidelity video—this skill allows developers to build sophisticated pipelines for generating, understanding, and iterating on visual and auditory content. It is a cornerstone for creators looking to leverage Openclaw Skills for automated media production.
Built on the official Google Gen AI SDK, this skill provides standardized patterns for handling binary outputs, managing large file uploads via the Files API, and orchestrating long-running asynchronous tasks like video synthesis. Whether you are building an automated social media engine or a complex data analysis tool that requires video understanding, this skill provides the technical foundation needed for robust implementation.
Gemini Multimodal Media (Image/Video/Speech) Use Cases
- Creating high-fidelity product images and cinematic videos from text descriptions using Nano Banana and Veo.
- Automating video analysis by extracting timestamped summaries and key events for research or compliance.
- Generating professional-grade narrations and podcasts with multi-speaker TTS and controllable vocal styles.
- Performing multi-image comparisons and visual QA to verify brand consistency or object detection.
- Building accessibility tools that transcribe audio and describe visual scenes for visually impaired users.
How Gemini Multimodal Media (Image/Video/Speech) Works
- Set up the environment by providing a valid Gemini API key and installing the required Node.js dependencies.
- Choose the specific capability needed—such as image understanding or video generation—and select the corresponding model from the Openclaw Skills matrix.
- Route media inputs through the Inline mode for small payloads or the Files API for large, reusable assets to optimize performance.
- Execute requests using the unified I/O specification, which handles different response modalities like TEXT, IMAGE, and AUDIO.
- Process binary outputs by decoding Base64 strings for images and PCM audio, or polling operation statuses for generated video files.
- Integrate outputs into downstream workflows, such as saving files to disk or passing generated data to another AI agent.
Gemini Multimodal Media (Image/Video/Speech) Setup
To get started with this skill, ensure you have Node.js 18+ installed. Install the official SDK and configure your environment variables:
npm install @google/genai
export GEMINI_API_KEY='your_api_key_here'
For REST-based implementations, ensure curl and jq are available in your environment for parsing JSON responses and decoding Base64 media data.
Gemini Multimodal Media (Image/Video/Speech) Data Schema & Taxonomy
The skill organizes data based on the modality and input method. Below is the primary metadata and I/O taxonomy:
| Capability | Model Recommendation | Output Format |
|---|---|---|
| Image Gen | gemini-2.5-flash-image | Base64 PNG/JPG |
| Image Analysis | gemini-3-flash-preview | Markdown/JSON Text |
| Video Gen | veo-3.1-generate-preview | MP4 (Async Download) |
| Video Analysis | gemini-3-flash-preview | Timestamped Text |
| Speech (TTS) | gemini-2.5-flash-preview-tts | PCM (24kHz, 16-bit) |
| Audio Analysis | gemini-3-flash-preview | Text Transcripts |
Gemini Multimodal Media (Image/Video/Speech) Advanced Features
- Multi-speaker TTS configuration allowing up to two distinct voices (e.g., Joe and Jane) for realistic dialogue generation.
- Advanced Veo 3.1 controls including aspect ratio adjustment (16:9/9:16), resolution settings up to 4K, and reference-image guidance.
- Intelligent Files API routing that automatically transitions from inline bytes to file uploads when exceeding a 20MB threshold.
- Compositional workflows that chain image generation with image understanding for automated self-correction and iterative editing.
- Support for native audio generation within video tasks, allowing for synchronized dialogue, ambience, and SFX.
SKILL.md
Loading
Related Openclaw Skills
Frontend Design Extractor
An automated tool for extracting and documenting UI/UX design systems from frontend codebases while stripping away business logic.
Wan Image Generation
A professional AI tool for generating high-resolution images via Alibaba's wan2.6-t2i model with automated local storage and cloud hosting.
OpenClaw Mem0 Plugin
A sophisticated memory integration that enables AI agents to remember user preferences, facts, and conversation history across multiple sessions.
MuskInsider Pro
A minimal HTTP service providing Elon Musk news briefings and demonstration payment integration for AI agents.
LiblibAI Image & Video Generation Skill
A comprehensive tool for generating professional AI images and videos using Seedream 4.5 and Kling models via the LiblibAI API.
Word Document Reader
A comprehensive command-line tool for parsing Microsoft Word documents and extracting text, tables, and metadata into structured formats.