Skill UI

This pattern addresses the fundamental limitation of context window size by treating the filesystem as a persistent, external memory. Agents can write large outputs, plans, or findings to files, offloading them from the active context. Instead of carrying all information, agents use targeted search and retrieval tools to load only the necessary context on demand, drastically reducing token bloat and improving performance over long, complex tasks.

This skill provides comprehensive tools for documenting and managing model performance on Hugging Face. It supports multiple methods for adding structured evaluation data, including extracting tables from README content, importing benchmark scores via APIs (like Artificial Analysis), and executing high-performance custom evaluations using advanced backends such as vLLM or lighteval. It ensures that all metadata is structured and compatible with model leaderboards and industry standards.

This skill formalizes successful, recurring patterns discovered in auto-memory (MEMORY.md) and graduates them into permanent, enforced rules within the project's core instruction set (CLAUDE.md or scoped rules). It ensures that learned best practices become mandatory project conventions, significantly improving adherence and consistency across all future AI interactions and development cycles. Use this when a temporary note proves to be a vital, repeatable guideline.

Achieve complete integration with Amazon Alexa to build sophisticated, intelligent voice skills. This system transforms a standard Alexa device into an ultra-intelligent assistant by utilizing Claude as the core LLM brain, while deeply integrating with the entire AWS ecosystem (Lambda, DynamoDB, Polly, Smart Home). Ideal for complex, specialized voice interaction applications.

This methodology converts verbose, natural language system instructions into an atomic, structured-label format (e.g., trigger:, action:). By structuring rules, it bypasses the limitations of natural language processing, ensuring LLMs allocate full attention to every constraint, thereby maximizing compliance, improving reliability, and reducing token usage.

Performs a full GEO audit of any website by discovering business type, crawling prioritized pages, and delegating parallel subagents to evaluate AI visibility, platform readiness, technical infrastructure, EEAT content, and schema markup, then outputs a composite GEO score with prioritized actions.

Generates and validates llms.txt files, the emerging standard that tells AI systems how to prioritize and understand a website; crawls content, ranks pages, and enforces formatting so your brand narrative stays consistent.

Framework for auditing and optimizing GEO content for each AI search platform, scoring Google AI Overviews, ChatGPT, Perplexity, Gemini, and Bing Copilot with tailored checklists and action steps.

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.

This skill orchestrates end-to-end autonomous AI research projects using a sophisticated two-loop architecture. It manages the entire research lifecycle, from initial literature review and hypothesis formation to running constrained experiments and synthesizing findings into publishable papers. It acts as a project manager, autonomously driving the process while structuring the workspace and presenting progress to human users.

This skill outlines the development of an automated pipeline to ingest raw Indicators of Compromise (IOCs) from various sources. It performs normalization, defanging (e.g., converting 'http://' to 'hxxp://'), and converts indicators into the standardized STIX 2.1 format. The processed, safe indicators can then be securely distributed via TAXII feeds, MISP, and other threat intelligence platforms.

This guide details comprehensive physical security penetration testing techniques, essential for red teaming engagements. It covers methods such as tailgating, badge cloning (RFID), lock bypassing, and deploying rogue network devices (e.g., Pineapple, LAN Turtle). It provides practical toolkits and MITRE ATT&CK mapping to evaluate an organization's physical and digital control weaknesses.