Skill UI

A comprehensive security tool designed to detect prompt injection attacks targeting Large Language Models (LLMs). It employs a multi-layered approach combining regex signature matching, structural anomaly scoring, and advanced DeBERTa transformer classification. It is essential for securing input validation layers in chatbots, AI agents, and RAG pipelines, and for retrospectively auditing LLM interaction logs against OWASP standards.

This comprehensive workflow provides full Create, Read, Update, and Delete (CRUD) capabilities for managing Miro boards and all embedded items. Users can programmatically manage entire boards, sticky notes, shapes, cards, frames, and tags using the official REST API v2. Ideal for integrating Miro into automated workflows, automating retrospective reports, or building complex collaboration tools that require structured board content manipulation.

This advanced tool surfaces critical, often unrecorded, evidence of human-AI collaboration during development sessions. It analyzes git diffs and conversational context across multiple layers to classify the signal level (HIGH/MEDIUM/LOW), measure technical complexity, detect decision forks, and quantify the specific contribution of the AI model, providing structured data for post-session retrospectives and accountability.