Analyze images with GLM-4.6V to generate captions and extract embedded text.
npx skills add https://github.com/archibate/archibate-skills --skill glm-vision
Please help me install this Agent Skill. Skill: glm-vision Source: https://github.com/archibate/archibate-skills/tree/main/optional-skills/glm-vision Command: npx skills add https://github.com/archibate/archibate-skills --skill glm-vision
Provides reliable visual understanding for user-submitted images by producing natural-language descriptions, extracting embedded text (OCR), identifying visual elements, and comparing multiple images to surface differences and semantics, which removes the need for manual inspection.
Please analyze the attached image, describe its main contents, extract any visible text, and list the key objects and colors present.
High-intent search queries and answers about installing and using this skill.
Image comparison analyzes multiple submitted photos or video frames to highlight differences. It identifies layout changes and similar objects across inputs using the multimodal vision API to surface semantic relationships.
Yes, GLM-4.6V supports analyzing batches of images and basic video frames. The skill applies the multimodal API to compare multiple inputs, extract embedded text, and generate natural-language descriptions.
Yes, the OpenAI library is required as a dependency to use Zhipu GLM-4.6V for image analysis. The skill includes basic error handling for missing API keys and processes Base64 or URL image inputs.
The best way to generate natural-language descriptions for photos is using a multimodal vision API. This skill produces concise summaries of visual elements, identifies key objects and colors, and extracts text without manual inspection.
Image analysis fails when the API key is missing because the Zhipu GLM-4.6V multimodal API requires authentication. The skill includes basic error handling to manage missing keys and prevent processing failures.
liumeixinConfigure custom vision backends for Hermes Agent with base64 image handling.
shekerkammaConvert code, docs, papers, images, and video files into queryable knowledge graphs.
LuminaVaultExtract text from WhatsApp images with Tesseract OCR and save to Obsidian.
TonyQ-AIAnalyze images with Xiaomi MiMo for OCR, UI review, and frontend debugging.
