Skill UI

Shortcut Automation lets you control Shortcut project management directly from Claude Code using natural-language commands to create or batch stories, list workflows and projects, add tasks or comments, and keep epics and workflows organized without leaving your terminal.

SimPO (Simple Preference Optimization) is a state-of-the-art, reference-free method designed for aligning Large Language Models (LLMs) using human preference data. It is an efficient alternative to DPO and PPO, notably outperforming DPO without requiring a separate reference model. It is ideal for practitioners who need faster, simpler, and more resource-efficient fine-tuning for preference alignment.

SAELens provides a framework for training and analyzing Sparse Autoencoders (SAEs). SAEs decompose the dense, often polysemantic activations of large language models into sparse, monosemantic features. Use this when you need to discover the discrete, interpretable concepts a model has learned, study feature superposition, or analyze specific safety-relevant behaviors (like bias or deception) within deep neural networks.

This guide details setting up a robust CI integration pipeline for Speak language learning platform integrations. It utilizes GitHub Actions with mocked API testing and audio validation to ensure code reliability and functional correctness before deployment. Ideal for developers managing complex language learning products and APIs.

This workflow provides an AI-powered platform for immersive conversation practice. It uses advanced language models (like GPT-4o) and real-time speech APIs to simulate real-life scenarios (e.g., ordering food, navigating). Users receive immediate, detailed feedback on pronunciation (phoneme level), grammar corrections, and vocabulary notes, making it an ideal core module for building language fluency and simulating professional tutoring experiences.

A comprehensive guide for optimizing costs when integrating the Speak API. This skill covers usage monitoring, selecting optimal API tiers, and implementing efficient audio processing pipelines. It is essential for developers managing language learning platform operations to ensure cost efficiency while maintaining high performance.

This skill provides a comprehensive guide for securely handling student audio data, assessment records, and learning progress within the Speak platform. It focuses on implementing robust data pipelines while ensuring compliance with international privacy regulations such as GDPR and COPPA. Essential for developers integrating or maintaining language learning applications.

A comprehensive diagnostic bundle designed for developers integrating with the Speak API. It systematically collects critical information, including authentication status, audio format validation (checking codecs, sample rates, and file size using ffprobe), and network connectivity tests for both Speak and OpenAI endpoints. Use this tool when troubleshooting complex language learning integration issues or verifying API setup.

This guide outlines the end-to-end process for deploying Speak language learning features to modern cloud environments, including Vercel, Cloud Run, or containerized setups. It covers initial configuration, implementing the core logic using TypeScript, and performing production verification. Use this skill when setting up a new Speak integration in a production environment or managing the platform's deployment lifecycle.

This skill guides the configuration of Speak for enterprise environments, such as schools and large organizations. It covers implementing Role-Based Access Control (RBAC), setting up Single Sign-On (SSO), managing diverse roles (teacher/student), handling class management, and establishing usage reporting. This setup is essential for deploying a secure, scalable language learning platform tailored for institutional use.

This comprehensive guide demonstrates how to build a robust AI language tutoring session using the Speak SDK. It covers initiating conversations, submitting student responses (text or audio), performing deep pronunciation assessment at the phoneme level, and generating detailed session summaries with grammar and vocabulary metrics. Ideal for developing language education platforms and testing API integration.

Provides a comprehensive incident response runbook for managing outages of the Speak API. It covers critical procedures including API triage, implementing offline fallback modes, and ensuring robust recovery steps. Ideal for DevOps teams maintaining language learning platforms reliant on real-time AI services (GPT-4o, Whisper).