Skill UI

A comprehensive guide and skill for deploying large language models (LLMs) locally using the llama.cpp framework and GGUF quantization format. It covers searching the Hugging Face Hub, selecting optimal quantizations (e.g., Q4_K_M), running local servers, and achieving OpenAI-compatible API endpoints. Supports various hardware backends including CPU, Mac Metal, CUDA, and ROCm. Ideal for developers needing offline, private LLM inference.

A comprehensive utility for building reusable, command-line scripts and utilities to interact with the Hugging Face API. It enables complex data workflows by allowing users to chain, pipe, and process data fetched from models, datasets, and various endpoints. Ideal for automation, enrichment, and advanced data pipelines.

The Neon AI Gateway provides a single API endpoint and one credential to access a diverse catalog of frontier and open-source LLMs (including OpenAI, Anthropic, Google Gemini, Meta, Alibaba, and DeepSeek). This unified layer simplifies integration by eliminating the need for separate provider accounts and managing model-specific SDK complexities. It allows developers to switch between models seamlessly by changing only the model name, while also providing crucial branch-scoped isolation for development and CI environments.

This skill automatically generates fully structured, import-ready Postman Collection v2.1 JSON files. It can interpret both natural language descriptions of REST APIs (endpoints, methods, parameters) and raw cURL commands. Use it whenever you need to quickly set up a comprehensive API testing suite in Postman without manual JSON construction, ensuring all variables and environments are properly extracted.

This skill provides expert capabilities for training and fine-tuning transformer language models using the TRL (Transformers Reinforcement Learning) library. It supports state-of-the-art post-training methods, including Supervised Fine-Tuning (SFT), Direct Preference Optimization (DPO), Group Relative Policy Optimization (GRPO), and Reward Model training. Use this skill to align and customize foundation models for advanced tasks via CLI commands.

This tool automatically generates comprehensive API documentation directly from source code, supporting TypeScript and JavaScript. It scans all public exports (functions, classes, interfaces) and extracts detailed JSDoc annotations. Furthermore, it can generate structured OpenAPI 3.0 specifications for HTTP endpoints, ensuring all public APIs are documented and compliant with industry standards.

This skill monitors for regressions on deployed URLs following deployments, merges, or dependency upgrades. It performs comprehensive checks including HTTP status, console errors, network failures, performance metrics (LCP/CLS/INP), and crucial API endpoint health, ensuring service stability in production or staging environments. Supports quick checks, continuous monitoring, and differential comparison against baselines.

This tool registers a Cognitum Seed IoT device by specifying its endpoint. It establishes a SeedClient connection, fetches the device's identity, and assigns an initial trust level. It supports optional pairing token usage for advanced setup and provides device status checks.

A comprehensive guide to Test-Driven Development (TDD) for Quarkus 3.x LTS, focusing on event-oriented architectures. This framework utilizes JUnit 5, Mockito, REST Assured, Apache Camel, and JaCoCo to ensure high code coverage (80%+). Use it when developing new features, fixing bugs, or refactoring event-driven services and API endpoints.

RouterBase acts as a unified, OpenAI-compatible API gateway, allowing applications to route requests across diverse large language models (LLMs) like GPT, Claude, and Gemini. It helps developers migrate existing OpenAI SDK calls, implement sophisticated primary/fallback model selection based on cost, latency, and quality, and manage multimodal payloads (image, audio, video) from a single, standardized endpoint.