Skill UI

CLIP enables general-purpose vision-language tasks through zero-shot image classification, similarity scoring, semantic search, and content moderation using pretrained contrastive embeddings without fine-tuning.

Gradient Clipping Helper automates ML training work by guiding gradient clipping workflows, generating production-ready code and configs, and validating outputs for PyTorch, TensorFlow, and related pipelines.

A comprehensive guide detailing Apple's Human Interface Guidelines (HIG) for integrating apps into system surfaces. Covers best practices for widgets, Live Activities, notifications, complications, App Clips, and other system components, ensuring apps provide glanceable, contextual utility outside the main app environment.

This skill enables the extension of existing video content using the Kling AI continuation model. By providing the task ID of a completed video and a descriptive prompt, it generates a seamless, continuous segment. This is ideal for content creators who need to build longer, multi-part narratives, expand short clips into full scenes, or generate complex, cohesive cinematic sequences by chaining multiple extensions.

LLaVA is an open-source vision-language assistant pairing CLIP vision encoders with Vicuna/LLaMA models to power multimodal chatbots, multi-turn image Q&A, captioning, and instruction-following tasks.

This skill guides the creation of structured, importable JSON templates for the Obsidian Web Clipper. It facilitates the analysis of a target website's content structure, including DOM, Schema.org data, and CSS selectors. Users can map complex web elements into a robust template format, incorporating advanced logic like conditionals and loops, ensuring the resulting clipboard data is consistently formatted and ready for immediate use.

Extract transcripts from YouTube videos, validate URLs and availability, then use STAR+R-I-S-E with youtube-transcript-api to craft comprehensive summaries and documentation for lectures, tutorials, and informational clips.

Formats Markdown into Slack-ready HTML or mrkdwn, previews the Slack-style rendering, copies rich HTML to the clipboard, and optionally delivers announcements via webhooks for messages and notifications.

Interactive workflow that interrogates scope, deliverables, constraints, and success criteria while guiding users to refine task prompts; copies the final markdown with Joyride clipboard commands and never writes code.

A comprehensive provider for integrating Eufy security cameras (supporting local P2P or cloud connectivity). This tool facilitates stable RTSP streaming for live monitoring and efficiently captures event-triggered video clips. It is designed for advanced systems that require continuous video feeds and reliable event-based data input for AI analysis and monitoring solutions.

This tool provides comprehensive integration for Reolink security cameras, supporting both RTSP streaming and HTTP API access. It allows users to discover cameras via ONVIF, capture live streams, take snapshots, and even capture motion-triggered video clips. It is ideal for developers building robust, networked surveillance or monitoring systems.

This skill provides comprehensive integration for TP-Link Tapo cameras (models like C200, C310, etc.). It supports streaming via RTSP and ONVIF protocols, allowing users to discover cameras on the network, capture motion-triggered video clips, and maintain stable live video feeds. Setup requires basic network credentials and Python environment setup.