Skill UI

This skill provides utilities to manage and drain the deliberation queue (brain_frames) of the ACC system. It allows users to list waiting frames, retrieve necessary context, perform deliberation on open tasks, and submit continuations using `acc_act(runtime="continue")`. It is essential for checkpointed work saved by headless runs, ensuring continuous task resolution and calibration against predicted outcomes in complex AI workflows.

This skill performs a comprehensive usability audit and heuristic evaluation of interactive user interfaces (UIs). It goes beyond mere visual aesthetics, assessing whether a system is intuitive, easy to learn, and functional for a first-time user. Use it when auditing complex tools, editors, or apps to ensure the user can complete primary tasks without needing extensive instructions or guidance.

Use this comprehensive skill for interacting with the Gemini API. It supports a wide range of advanced AI tasks, including text generation, multi-turn chat, multimodal understanding, image/video creation, function calling, and structured output. Provides modern SDKs for robust application development in Python and JavaScript/TypeScript.

A comprehensive guide for developers building AI demos and GPU compute tasks on Hugging Face Spaces using ZeroGPU hardware. It covers best practices for writing and reviewing code that utilizes the `@spaces.GPU` decorator, managing runtime duration and quota, and understanding the specific CUDA availability model and process isolation rules required for reliable ML deployment.

A comprehensive guide for training or fine-tuning advanced sentence-transformer models. This skill assists in selecting the correct architecture—including Bi-encoders (SentenceTransformer), CrossEncoders (for reranking), and SparseEncoders—for various Natural Language Processing (NLP) tasks. Use cases include semantic retrieval, similarity matching, document clustering, and pair-wise classification.

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 meta-skill acts as a routing layer for complex UI development tasks. It guides the process of selecting the smallest, most specific, and useful skill context using the ui-skills CLI. Use it when starting UI work to analyze goals, identify the correct category, and pinpoint the precise skill set needed for implementation, ensuring development efficiency and accuracy.

This template provides a comprehensive three-screen UI showcase for a gamified habit tracking application, designed with an RPG (Role-Playing Game) aesthetic. It includes a main cover page, a 'Today's Quests' summary screen displaying XP and level progress, and a detailed quest view for managing subtasks and rewards. Perfect for showcasing personal growth, productivity, or educational concepts.

A comprehensive Notion dashboard designed for real-time team operations and performance tracking. It provides a centralized overview by aggregating Key Performance Indicators (KPIs) in a grid format, displaying 7-day trend sparklines, showing live activity feeds, and summarizing tasks in a structured database table. Ideal for project managers and team leads who need a single source of truth for project status and team coordination.

This skill utilizes the built-in macOS `caffeinate` command to prevent the Mac from entering sleep mode. It is essential for guaranteeing uninterrupted workflow during long-running tasks, such as code builds, large downloads, or supervised automation sessions. Users can specify a fixed duration or attach the function to a specific process ID for continuous wakefulness.

This skill executes the DeepSWE benchmark, a comprehensive coding-agent evaluation suite. It uses OpenRouter and mini-swe-agent to score various Large Language Models (LLMs) on 113 tasks, providing reproducible results for model comparison, development assessment, and research benchmarking. Use this when a standardized, deep evaluation of LLM coding capabilities is required.

This skill guides users through complex, multi-step processes such as installation, configuration, or initial setup. It breaks down the process into small, atomic actions, presenting only the current required step and a limited, manageable checklist of remaining tasks. Ideal for onboarding new users or executing lengthy technical setups.