Skill UI

Provides LLM coding agents with a persistent graph of code entities, imports, public APIs, and dependency reasons, enabling navigation, refactoring, and impact analysis without rereading the whole repository.

Build, maintain, and extend the EarLLM One Android app, a Kotlin and Compose voice pipeline that connects Bluetooth earbuds to an LLM.

This skill enables the creation of robust evaluation frameworks for AI agent systems. It is used for systematically testing agent performance, validating context engineering choices, and measuring improvements over time. The approach accounts for non-determinism and multiple valid paths, utilizing multi-dimensional rubrics, LLM-as-judge methods, and human evaluation to ensure quality gates and catch regressions before deployment.

Extract keywords with YAKE or RAKE, detect document language, and generate vector embeddings for RAG and search using the xberg CLI or Python API, with local, LLM, or plugin providers.

Extract structured tabular data from PDFs, spreadsheets, and images using layout-aware detection and configurable table models, with markdown or JSON cell output for LLM ingestion and downstream processing.

Analyzes and generates llms.txt files, the emerging standard that gives AI systems a structured, machine-readable summary of a website's purpose, key pages, and facts. Covers format rules, section conventions, and the extended llms-full.txt variant for AI discoverability.

Estimate the required memory for inference of Safetensors or GGUF models from the Hugging Face Hub using HTTP Range requests, without downloading weights. Supports optional KV cache estimation for LLMs and VLMs.

Run evaluations for Hugging Face Hub models using inspect-ai and lighteval on local hardware. Handles backend selection, local GPU evals, and choosing between vLLM, Transformers, and accelerate.

This skill enables users to add and manage structured evaluation results within Hugging Face model cards. It supports extracting evaluation tables from README files, importing benchmark scores via the Artificial Analysis API, and executing custom model evaluations using vLLM or lighteval backends. It ensures compatibility with the model-index metadata format for standardized model releases.

This skill guides users on running evaluations against models on the Hugging Face Hub using local hardware. It covers using inspect-ai and lighteval frameworks with local inference backends like vllm, Transformers, or accelerate. It includes task selection, smoke testing, and hardware guidance for different model sizes. It does not cover remote orchestration or publishing results to the community workflow.

This skill focuses on identifying security vulnerabilities in Large Language Model (LLM) and AI features. It covers prompt injection, system prompt leakage, cross-tenant access, and data exfiltration techniques. Users must verify findings against false positives using reproducibility checks and out-of-band callbacks before reporting.

Master the LangChain framework to build sophisticated LLM applications. Learn to implement agents, chains, memory management, and tool integration. This skill covers core concepts like ReAct agents, retrieval chains, and document processing pipelines. Ideal for creating autonomous AI agents and production-grade applications using Python.