Browse
Meta
Skills Found
architect-agent
Coordinates planning and evaluation between architect and code agent workspaces. Provides clear intent classification, file location protocols, and workflow guides for instruction creation, grading, and workspace setup. Uses template-based initialization for quick deployment.
interviewing-user
A well-structured skill for systematic requirements gathering that proactively interviews users to resolve ambiguities before implementation, preventing wasted effort from incorrect assumptions.
multi-agent-coordination
Provides concrete patterns for coordinating multiple Claude agents in workflows, focusing on parallel vs sequential execution. Explains the 4-message pattern for true parallelism and file-based delegation to avoid context pollution. Includes agent selection logic and anti-pattern warnings.
using-ring
A mandatory orchestration skill that enforces a 'dispatch agents, don't operate directly' principle. It implements a strict 3-file rule for direct tool use, requires pre-action checkpoints before any operation, and mandates automatic agent dispatch for multi-file tasks. Includes context preservation protocols with mandatory handoff triggers at 70% and 85% context usage.
project-astrology-tarot-divination
This skill provides a structured framework for Tarot card readings, including card meanings for all 78 cards (upright and reversed), common spread layouts like the Celtic Cross, and integration with astrological timing. It enforces ethical guidelines for readings and includes decision trees to route user requests to appropriate resources.
.claude
A repository management skill that provides Claude with context about a prompt library's structure and tools. It explains file naming conventions, directory organization, and Python utilities for maintaining a table of contents and generating GPT documentation templates.
compound-learnings
Analyzes session learnings to identify recurring patterns across 3+ sessions, then proposes converting them into rules, skills, or hooks. Requires explicit user approval before creating any artifacts. Helps teams systematically capture and institutionalize successful practices from daily work.
mastra
Mastra is an agent framework that orchestrates multiple tools and services to complete complex tasks. It provides structured workflows for handling multi-step processes, with built-in error handling and tool coordination. The framework focuses on reliability and clear execution paths.
smart-routing
An intelligent routing engine for Claude's /toh command that analyzes user intent, calculates confidence scores, checks memory context, and routes requests to appropriate specialized agents. It handles ambiguous inputs, detects IDE environments for parallel execution, and maintains project context through memory files.
discover-math
A gateway skill that auto-detects math-related tasks and provides access to 11 specialized math skills. It uses a progressive loading system where users first see the gateway, then can browse the full index, and finally load specific skills as needed. The skill activates automatically when Claude detects mathematics, algorithms, or related topics.
moai-foundation-context
Manages Claude Code's 200K token context window with budget tracking, session persistence, and multi-agent handoffs. Provides concrete patterns for when to clear context, how to allocate tokens, and ensures workflow continuity across model switches.
project-astrology-chinese
This skill provides detailed information about the Chinese zodiac system, including the 12 animal signs, Five Elements (Wu Xing), BaZi (Four Pillars) birth chart analysis, and compatibility guides. It functions as a structured reference tool for answering specific questions about Chinese astrology concepts, calculations, and interpretations.