Agent Skills · AI Agent
Curated skills for Claude, Cursor and other AI agents. Copy one prompt to install.
An optimization loop for existing Skills that improves SKILL.md through nine-dimension scoring, blind judging, prompt testing, and automatic rollback.
Provides AI agents with a remote Markdown profile to recognize user identity and preferences, supporting reads, updates, and cross-tool sharing.
Turns informal notes and meeting records into structured documents, task lists, and weekly reports, and checks procurement, reimbursement, and hazardous-materials workflow deliverables.
Distills successfully debugged workflows into reusable Agent Skills, preserving verified commands, failure modes, and excluded paths.
An AI agent that routes intent, runs a one-pass 5W1H intake, and produces generated, analyzed, planned, or query-style results with safety and format self-checks.
Research a named person or fuzzy need through three sources, distill a thinking framework, and generate a runnable Skill for optimization and release.
Guides creation, initialization, editing, and packaging of reusable AI agent skills with directory patterns, progressive loading, and best practices.
Use dLazy CLI to call Grok 4.2 for text generation, QA, and reasoning.
A Java-based prompt orchestration skill that turns vague requests into executable instructions with multi-model, RAG, and local-cache support.
Search, browse, and install skills from the open agent skills ecosystem to quickly find capabilities by task.
Engine-neutral NPC behavior design covering FSMs, behavior trees, pathfinding, and steering for decoupled decision, routing, and motion.
A local Python prompt assistant that configures multiple model APIs for requirement clarification, prompt generation, document generation, and local caching.