Skill UI

Guides researchers through nnsight’s tracing and activation APIs so you can inspect or modify PyTorch internals locally or via NDIF-powered remote runs on 70B+ foundation models without local GPUs.

Runs build, test, lint, serve, and other tasks in Nx workspaces via run, run-many, and affected flows, honoring the detected package manager and target metadata while exposing filters, parallelism, and caching controls for CI or local workflows.

This guide establishes a comprehensive and rapid development loop for Obsidian plugins. It covers the entire workflow, from setup (cloning templates, symlinking) to advanced features, including running esbuild in watch mode, achieving hot-reloading (via Ctrl+R or automatic methods), deep debugging with Chrome DevTools, and running unit tests with vitest. Essential for developers seeking fast feedback and reliable plugin iteration.

This skill automates the setup and configuration of Ollama, allowing users to deploy Large Language Models (LLMs) locally. It performs system assessment (OS, GPU, RAM), selects optimal models based on hardware availability, and handles installation across macOS, Linux, and Docker. Use this when you require free, self-hosted AI alternatives, need offline inference capabilities, or wish to eliminate reliance on costly hosted API services.

This provides a local development loop for integrating the OpenEvidence clinical decision support API. Developers can build health-tech tools and quickly iterate using mock evidence queries, citation responses, and clinical summaries. It supports toggling between mock mode for rapid development and sandbox mode for validation against the real platform, ensuring development uses de-identified data.

Outlines guarantees valid JSON/XML/code outputs by applying CFG/FSM-constrained token filtering, Pydantic schemas, and grammar-level sampling while supporting local Transformers, llama.cpp, and vLLM backends for fast inference.

Run local Lighthouse audits via CLI or Node API, parse JSON/HTML reports, enforce performance budgets, integrate with CI (Actions/LHCI), and compare builds to catch regressions before deploys.

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.

Guides building productivity-first tools by scratching your own itch—rapid prototyping, CLI/local-first apps, scripting to product evolution, and dogfooding for reliable personal workflows.

This guide details the recommended pattern for managing per-project plugin settings and state within a development environment. Plugins store their configuration in a dedicated `.claude/plugin-name.local.md` file. This file utilizes YAML frontmatter for structured settings (like enabling flags, retry counts, and parameters) and a Markdown body for additional context or instructions. It provides techniques for reading these settings using Bash hooks, commands, and agents.

This guide provides a comprehensive workflow for setting up a robust local development environment for PostHog integrations. It covers enabling debug mode for real-time event inspection, implementing mocking strategies for stable unit testing, and configuring integration tests using a dedicated development PostHog project. Use this when building or testing analytics tracking features with posthog-js or posthog-node to ensure fast, isolated development cycles.

This comprehensive guide outlines advanced strategies for optimizing PostHog in production environments. Key techniques covered include implementing local feature flag evaluation to eliminate network latency, optimizing event batching for network efficiency, utilizing strategic event sampling, and writing efficient HogQL queries. Master these methods to dramatically improve data collection reliability and query speed.