Agent Skills Hub

web-search-patterns

Advanced Gemini-powered web search plugin with smart caching, subagent context isolation, and automated query optimization.

parallel-agents

Orchestrate complex workflows by coordinating multiple specialized AI agents for multi-perspective code analysis, feature implementation, and system-wide reviews.

symbols

Token-efficient codebase navigation through intelligent symbol indexing, domain chunking, and architectural layer filtering. Reduce token usage by 60-95% when exploring or developing complex systems.

service-mesh-observability

Implement production-grade observability for Istio and Linkerd service meshes, including distributed tracing, metric dashboards, and golden signal monitoring.

higress-auto-router

Manage automatic model routing for Higress AI Gateway via CLI. Configure triggers for intelligent model selection based on request content.

new-analyzer

Guide for implementing a new AI coding agent analyzer in Splitrail to track token usage, costs, and performance metrics.

everything-claude-code-cn

A comprehensive configuration suite for Claude Code, featuring production-grade agents, skills, hooks, and automated workflows optimized for high-intensity development.

getting-started-guide

A structured guide for novelists to navigate the seven-step writing process, from constitution and specification to planning, tasking, drafting, and quality analysis.

deepwiki-rs

AI-powered documentation engine that automatically generates C4 architecture diagrams, technical specs, and codebase analysis from any source code directory.

basic-usage

Essential guide to llmemory for document storage and search: installation, database setup with pgvector, document ingestion, hybrid/semantic retrieval, and building RAG systems with multi-tenant support.

context-fundamentals

Foundational guidelines for context engineering: optimizing token budgets, attention mechanics, and system architecture for AI agents.

search

Search the web using Tavily's LLM-optimized search API for relevant, source-cited content without writing code.