Skill UI

This skill provides a detailed, side-by-side comparison of advanced LLM agent evaluation frameworks, such as ruflo and HAL. It helps researchers and developers identify capability gaps—including real code execution, full web browsing, and structured file parsing—and prioritize architectural improvements for robust agent benchmarking and performance tracking.

This comprehensive guide provides authors with best practices for preparing supplementary artifacts for AAAI submissions. It outlines how to structure technical appendices, package reproducible code and data ZIPs, maintain strict anonymity (by removing author metadata and paths), and maximize the evidence for reproducibility. The goal is to ensure that external reviewers can easily verify the research claims without specialized or private tools.

This skill converts unstructured documents (PDF, EPUB, DOCX, etc.) into actionable, structured agent skills. Instead of providing simple summaries, it deep-dives to extract core frameworks, mental models, key principles, techniques, and anti-patterns from educational material. Use this when you need to transform academic knowledge into a reusable, machine-readable toolkit for agents like Copilot CLI, Amp, or Claude Code.

This skill extracts crucial Social Determinants of Health (SDOH) — such as housing instability, food insecurity, and transportation barriers — from unstructured clinical notes. It maps the identified text spans to standardized ICD-10-CM Z-codes (Z55–Z65). It is essential for health-equity analytics, building risk stratification tools, and identifying coding gaps by converting narrative text into queryable, discrete data points.

This technique segments unstructured clinical documents (such as HPI, PMH, and A&P sections) into contextually rich, canonical chunks. This is crucial for improving the precision of downstream NLP tasks like Named Entity Recognition (NER) and de-identification. By identifying and normalizing headers and mapping them to LOINC codes, the system ensures that the source context (e.g., historical vs. active) is maintained, leading to significantly more accurate results.

This skill processes verbose, stream-of-consciousness voice dictations (from tools like macOS Dictation or Wispr Flow) and transforms them into highly structured, token-efficient prompts. It removes filler words, repetitions, and tangents, organizing the core intent into standard sections (Context, Objective, Constraints) suitable for advanced LLMs like Claude Code.

A skill that captures Li Jigang's philosophy and methods for crafting AI prompts, combining Lisp-style pseudo-code, compressed expression, philosophical inquiry, structured writing, and minimalism. Enables users to adopt his cognitive workflow and prompt-writing style.

This skill structures a highly controlled, bounded development life cycle for AI agents. It mandates a structured three-phase workflow: Planning (Spec), Implementation (Build), and Verification (Review). It is essential for building complex features from scratch, especially when dealing with ambiguous requirements, high risk, or needing explicit human approval gates to ensure the resulting code meets all defined specifications.

Convert human-written CLAUDE.md into AI-native structured-label format. Battle-tested across 4 models, reducing token usage while increasing compliance. Ideal for optimizing LLM system prompts and migrating rules between AI tools.

This skill provides structured patterns for combining vector similarity search with traditional keyword-based search. It is essential for building advanced Retrieval-Augmented Generation (RAG) systems, sophisticated search engines, or any retrieval task requiring high recall. Use this when pure semantic understanding is insufficient, and precise matching of specific terms (like names or codes) is necessary.

Lambda is a compact, machine-native communication protocol specifically designed for agent-to-agent (A2A) messaging. It functions as a shared, unambiguous vocabulary built from 'atoms' (2-character codes), completely bypassing the need for human natural language interpretation. It is ideal for orchestrators, task delegation, and logging structured coordination signals (like heartbeats or state changes) in multi-agent systems. Crucially, Lambda is intended only for machine communication, not for user-facing dialogue.

Design and optimize production-grade multi-agent systems using LangGraph, LangChain, and DeepAgents. Provides structured workflows for creating supervisor, planner, researcher, coder, and validator agents with memory and tool-calling pipelines.