Skill UI

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.

Enables Trigger.dev-based background jobs, AI workflows, and scheduled or batch async execution with a TypeScript-first SDK, durable retries, and built-in integrations while matching production behavior through the local CLI.

An expert guide for deploying and optimizing Large Language Models (LLMs) entirely on local hardware. Mastering the entire local AI ecosystem, including inference engines like Ollama, vLLM, and llama.cpp. Provides deep knowledge on quantization formats (GGUF, EXL2, AWQ), VRAM calculation, and ensuring privacy-first, offline deployment for state-of-the-art models.

Utilize Modal, a serverless GPU cloud, for running complex machine learning workloads, including training, fine-tuning, and inference. This tool provides a zero-config environment, eliminating the need for SSH setup or Docker management. It features automatic scale-to-zero billing and local-first execution, making high-powered GPU compute accessible for debugging and small-scale experiments.

This structured skill guides users through root-cause-first debugging when an implementer is blocked, verification fails, or repeated attempts at fixing do not converge. It mandates a systematic approach combining local evidence inspection, runtime analysis, and external documentation research to accurately classify the failure's true source (e.g., missing dependency, spec conflict). It prevents speculative patching and generates a clear, actionable fix plan.

This skill provides implementation patterns for integrating with RemoFirst, a comprehensive global platform for HR, EOR, and payroll management. Use this when developing workflows that require accessing global employee data, managing cross-border payroll, or ensuring international compliance. It guides API interaction and local development loops for global workforce management.

This skill enables the creation and registration of reusable cloned voices from local audio samples for subsequent Text-to-Speech (TTS) applications. It implements a strict 'Consent-first' workflow, requiring explicit user consent, detailed metadata (speaker, permitted use), and adherence to copyright laws before any voice cloning can occur. It ensures the resulting voice ID is safe, traceable, and legally compliant for professional content creation.

Pilot Protocol is an open-source overlay network providing AI agents with first-class network citizenship. It offers permanent virtual addresses, encrypted P2P communication, NAT traversal, and a structured app store. Use this skill when agents require stable networking across cloud changes, need direct, secure messaging without shared accounts, or must integrate structured external data services (e.g., live crypto prices) via local, typed capabilities.

Tree Ring Memory is a local-first, framework-agnostic layer for managing durable knowledge in AI agents. It provides structured methods for recalling past decisions, recording evidence, auditing project state, and deliberately forgetting information, ensuring the core project memory remains clean and actionable without relying on raw conversation transcripts.

This skill offers two complementary workflows for locally checked-out codebases. Documentation mode creates a definitive, cited architecture document ideal for onboarding and system research. Modernization mode then builds upon this documentation, generating a phased, safety-laddered plan to upgrade, migrate, or rewrite legacy systems. It emphasizes local-first analysis for maximum accuracy.

This skill acts as a comprehensive map for developing clinical and biomedical NLP pipelines using OpenMed. OpenMed is a local-first Python library that performs named-entity recognition (NER), protected health information (PHI) de-identification, and FHIR export, all running fully on-device without cloud calls. Use this guide when starting a new clinical text project, needing to select the appropriate capability (e.g., zero-shot NER, auditing, or FHIR conversion), or requiring an entry point for the core OpenMed features.

Guideline for front-end JavaScript reverse engineering using js-reverse MCP tools and jshookmcp. Covers locating signature and encrypted parameters, tracing request initiators, runtime sampling via hooks, local Node environment reconstruction, and evidence-first output for reproducing signing logic.