An advanced AI tool for deep understanding of images, videos, and complex documents using the GLM-4.6V model.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install multimodal
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 multimodal 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 Multimodal Analyzer?
The Multimodal Analyzer is a powerful addition to your Openclaw Skills library, designed to bridge the gap between text-only AI and visual data. Powered by the high-performance GLM-4.6V model, which features 106B parameters and a 128K context window, it provides state-of-the-art vision capabilities including high-accuracy OCR, object detection, and comprehensive video summarization.
This skill enables developers to build smarter agents capable of interpreting real-world visual content, complex PDF tables, and long-form video files. By integrating this into your Openclaw Skills ecosystem, you unlock sophisticated reasoning that goes beyond simple image recognition, allowing for deep contextual analysis across multiple media types.
Multimodal Analyzer Use Cases
- Extracting structured data and insights from complex PDF tables and technical documents.
- Generating descriptive summaries and keyframe analysis for long-form video content.
- Performing high-accuracy OCR and scene analysis for automated visual inspection tasks.
- Integrating deep reasoning into image-based workflows using the specialized thinking mode.
How Multimodal Analyzer Works
- Initialize the skill by configuring the ZHIPU_API_KEY environment variable to access the GLM-4.6V model.
- Provide a target resource via a publicly accessible URL or a local file path.
- Specify the modality type (image, video, or file) to guide the processing engine.
- The skill transmits the data to the GLM-4.6V endpoint for multimodal interpretation.
- The system returns a structured analysis or a real-time streamed response based on your specific prompts.
Multimodal Analyzer Setup
To use this within your Openclaw Skills environment, ensure you have the necessary environment variables configured.
# Set your Zhipu AI API Key
export ZHIPU_API_KEY="your_api_key_here"
# Execute an image analysis
python scripts/analyze.py --type image --input "./photo.jpg" --prompt "Identify all objects in this image."
# Execute a video analysis with thinking mode enabled
python scripts/analyze.py --type video --input "https://example.com/video.mp4" --prompt "Summarize the main events." --thinking
Multimodal Analyzer Data Schema & Taxonomy
The Multimodal Analyzer processes inputs based on the following schema parameters:
| Parameter | Type | Description |
|---|---|---|
--type |
String | Defines input modality: image, video, or file (PDF). |
--input |
String | Local path or public URL of the asset. |
--prompt |
String | The natural language instruction for analysis. |
--thinking |
Flag | Enables deep reasoning mode for complex queries. |
--stream |
Flag | Enables token-by-token streaming response. |
Multimodal Analyzer Advanced Features
- Deep Thinking Mode for multi-step logical reasoning over visual and document data.
- Large Context Support allowing for the analysis of lengthy documents and extended video clips.
- Streaming Output capability for building responsive user interfaces within Openclaw Skills.
- Comprehensive OCR and Table Extraction for digitizing complex paper-based workflows.
SKILL.md
Loading
Related Openclaw Skills
Overseerr Request Media
A specialized tool for automating movie and TV show requests through your personal Overseerr instance using natural language.
Discord Shopping Deals
A specialized shopping agent that scouts major retailers and refurbished marketplaces to deliver the top 3 best deals directly into Discord channels.
Catholic Grounding
A specialized skill for providing accurate, Catechism-cited Catholic teaching and spiritual resources via AI agents.
OpenExec
A security-focused execution service that ensures only pre-approved, deterministic actions are processed with verifiable cryptographic receipts.
Trinity Compress
A local script that compresses prompt files to minimize token costs in iterative AI development cycles.
Trinity Compress
A lightweight local script that compresses prompt files to minimize token usage in iterative AI development cycles.
Featured*
