Skill UI

High-performance RLHF framework that combines Ray, vLLM, and ZeRO-3 to accelerate PPO/GRPO/RLOO/DPO training for 7B-70B+ models on distributed GPU clusters, providing a single-stack workflow for reward models and policy optimization.

Enables Claude to orchestrate complex test workflows via the test-orchestrator plugin, defining dependencies, parallel execution, and smart selection so only affected suites run during CI/CD or manual optimization requests.

Automates load, stress, spike, and endurance tests through the performance-test-suite plugin, collecting response time, throughput, and error-rate metrics to spot CPU/memory/database/network bottlenecks and deliver graphs plus optimization guidance.

A comprehensive guide and implementation for setting up a fast, cost-effective local development loop for the Perplexity Sonar API. It includes generating type-safe client wrappers, utilizing fixtures for reliable mocking, and establishing a complete test suite with hot-reloading capabilities. This ensures stable development and manages high API call costs during iteration.

This skill specializes in designing, optimizing, and evaluating prompts for Large Language Models (LLMs). It allows users to generate optimized prompt templates, structured output schemas (e.g., JSON mode, function calling), and comprehensive test suites. Use this skill when building new LLM applications, implementing complex reasoning patterns like Chain-of-Thought, defining system guardrails, or building robust evaluation frameworks to measure and improve model performance.

A comprehensive skill set for building robust, type-safe, and highly performant Python 3.11+ applications. This specialist enforces best practices by generating type-annotated code, utilizing async/await patterns, structured dataclasses, and comprehensive error handling. The workflow mandates rigorous validation using tools like mypy (strict mode), black, ruff, and extensive pytest test suites to ensure code quality and reliability.

Provides a comprehensive pytest-focused testing strategy for Python applications, covering TDD cycles, fixtures, parametrization, coverage tools, and test infrastructure setup for reliable suites.

This comprehensive guide provides robust strategies for testing Python code using industry best practices. It covers essential techniques like pytest, fixtures, mocking, parameterization, and implementing Test-Driven Development (TDD). Use this skill when writing unit tests, setting up complex test suites, performing integration testing, or establishing continuous testing infrastructure in CI/CD pipelines.

Centralizes pipeline health checks, forecast accuracy tracking, and go-to-market efficiency measurement for SaaS revenue teams, generating text or JSON outputs via CLI scripts to monitor coverage, MAPE, GTM metrics, and risk flags before leadership reviews.

Automates environment prep, service orchestration, and cleanup for executing integration test suites, covering database setup, migrations, and optional coverage reports.

Introduces deliberate code mutations, reruns the suite, and reports surviving mutants plus a mutation score whenever the user asks to assess or improve test effectiveness.

This skill provides a comprehensive suite for executing load testing, stress testing, and performance benchmarking. Use it to simulate peak traffic, identify system bottlenecks, and verify compliance with Service Level Agreements (SLAs). It supports industry-standard tools like k6, Artillery, Locust, and JMeter for robust performance analysis.