Skill UI

Stable Baselines3 is a PyTorch-based library providing production-ready implementations of core reinforcement learning algorithms, including PPO, SAC, DQN, and TD3. It offers a scikit-learn-like API for easily training RL agents, developing custom Gymnasium environments, and managing complex training workflows. Ideal for standard RL research, rapid prototyping, and advanced model optimization.

A comprehensive security auditing framework designed to systematically detect sensitive information disclosure from Large Language Models (LLMs). It uses multi-phase, escalating dialogue probes to test for leaks such as system prompts, API keys, PII, and internal configurations. Essential for red teaming and rigorous model security assessment.

Detects goal hijacking vulnerabilities where malicious instructions are hidden within external content sources processed by the agent, such as retrieved documents (RAG), uploaded files, or web page content. This skill simulates these attack vectors to ensure the agent only uses the provided context for the stated task and does not execute embedded hidden commands.

This comprehensive skill provides multi-layered guardrails for integrating with LLMs like Claude. It enforces security and compliance by implementing input validation (PII detection, length checks), system prompt restriction (scope control, safety rules), output filtering (preventing sensitive data leakage), cost governance, and model access policies. Ideal for building secure, robust, and governed AI applications.

This skill provides comprehensive guides for deploying stateless, serverless API wrappers powered by Claude (Anthropic). It details how to integrate and stream responses from the Messages API across major cloud environments, including Vercel Edge Functions, Fly.io, and Google Cloud Run. It is essential for productionizing AI applications that require robust secret management, scalable infrastructure, and real-time streaming capabilities.

A comprehensive checklist designed to ensure Claude-powered applications are robust, secure, and production-ready before going live. It guides developers through critical areas, including robust error handling, API security best practices, cost control mechanisms, real-time monitoring setup, and performance optimization, ensuring a stable user experience.

This skill provides comprehensive deployment guides for Cohere API v2 applications across major cloud platforms including Vercel, Fly.io, and Google Cloud Run. It covers critical aspects like secret management, setting up reliable health checks, optimizing performance, and implementing streaming chat endpoints to ensure production readiness.

This guide provides comprehensive instructions and code examples for deploying ElevenLabs Text-to-Speech (TTS) applications to various cloud environments, including Vercel (serverless), Fly.io (containers), and Google Cloud Run. It covers best practices for managing platform-specific secrets, configuring timeouts, and implementing real-time streaming audio capabilities for production-grade voice applications.

This benchmark evaluates Vision-Language Models (VLMs) on detecting potential safety hazards within indoor environments. It uses static camera frames, simulating real-world fixed security camera monitoring. The test covers 40 scenarios across 5 critical categories, including fire/smoke, electrical risks, trips/falls, child safety, and falling objects, providing a rigorous measure of VLM's practical safety inspection capability.

This harness provides comprehensive, reproducible evaluation pipelines for complex LLM chains and agents (LangChain/LangGraph 1.0). It integrates golden dataset management, LangSmith evaluation, RAGAS metrics, deepeval LLM-as-judge, and structured agent trajectory analysis. Use it when establishing quality benchmarks for new chains, diagnosing performance regressions after model switches, or implementing CI/CD gates to prevent quality drops.

This skill translates complex legal jargon (legalese) from contracts and agreements into clear, plain English at an 8th-grade reading level. It systematically identifies and rewrites material clauses, replacing archaic terms with everyday equivalents. Crucially, it also flags potentially misleading or high-risk sections, providing a clear, non-technical 'What You Are Actually Agreeing To' summary for non-lawyers.

This skill performs rigorous mathematical proof verification using an adversarial, cross-model review process. It reads a LaTeX proof, systematically identifies logical gaps (e.g., unjustified assertions, quantifier errors) and theoretical flaws (e.g., illegal interchanges, non-uniform convergence). It then iteratively fixes these gaps with full derivations, re-reviews, and generates a comprehensive audit report, ensuring mathematical rigor and completeness. Use this when seeking deep, academic verification of a theory paper.