Claude Skills

This skill enables semantic vector search using AgentDB for building RAG systems and knowledge bases. It provides high-performance document retrieval with HNSW indexing and sub-millisecond query speeds. Use it for intelligent document retrieval, similarity matching, and context-aware querying in your applications.

This skill optimizes AgentDB performance by implementing quantization for 4-32x memory reduction and HNSW indexing for up to 150x faster vector searches. It also provides caching strategies and batch operations to scale databases to millions of vectors. Use it when you need to improve search speed, reduce memory usage, or handle large-scale vector data efficiently.

This skill provides persistent memory patterns for AI agents using AgentDB, enabling session memory, long-term storage, and context management. Use it when building stateful agents, chat systems, or intelligent assistants that need to remember conversations and learn from interactions. It offers significant performance improvements over traditional solutions while maintaining full backward compatibility.

AgentDB Learning Plugins provides a suite of 9 reinforcement learning algorithms, including Decision Transformer and Q-Learning, for building and training self-learning AI agents. It enables developers to create plugins that optimize agent behavior through experience, with WASM-accelerated training for faster performance. Use this skill when implementing reinforcement learning or developing autonomous agents that need to improve over time.

This skill teaches advanced AgentDB features for building distributed AI systems, including QUIC synchronization for sub-millisecond cross-node communication and hybrid vector search. Use it when developing applications requiring multi-agent coordination, custom distance metrics, or multi-database management. It's designed for developers implementing production-scale, distributed vector search systems.

This skill performs automated accessibility testing for WCAG 2.2 compliance, including screen reader validation and inclusive design checks. Use it to ensure legal standards (ADA, Section 508) are met and to build applications accessible to disabled users. It applies POUR principles, tests keyboard navigation, validates with screen readers, and checks color contrast.

The case-study-writer skill helps developers create structured blog posts to share AI/automation project results internally. It interactively collects project details (problem, solution, results) and generates formatted case studies in markdown. Use this skill after completing a project to document and communicate its impact in a consistent, company-appropriate format.

This Claude Skill performs [specific task] through a defined step-by-step process. It takes required and optional inputs from the user to generate structured outputs, making it useful for [primary use case]. Developers can integrate it via the `/스킬이름` command for automated task execution.

This skill helps developers create case study blog posts for sharing AI/automation project results internally. It interactively gathers project details (problem, solution, outcome) and structures them into a ready-to-publish markdown format. Use it after completing a project to document and communicate the implementation's impact within your organization.

This Claude Skill performs [스킬이름] to [목적]. It works by [동작 방식 단계 요약] and outputs [출력 형태]. Use it when you need to [언제 사용하면 좋은지].

This skill generates comprehensive API documentation from source code, including OpenAPI specifications and endpoint details. It's ideal for creating or updating API docs when users mention endpoints or documentation needs. The tool produces structured documentation with parameters, responses, and code examples for each endpoint.

This Claude Skill performs comprehensive code reviews with security, performance, and quality analysis. Use it to evaluate code, pull requests, or for security and optimization tasks. It checks for vulnerabilities, algorithm efficiency, SOLID principles, and maintainability issues.

This skill ensures all generated text adheres to defined brand voice, tone, and messaging guidelines. It is designed for creating marketing copy, customer communications, and any public-facing content. Developers can use it to automatically maintain brand consistency in AI-generated text.

This skill diagnoses and resolves common technical issues in Claude Patent Creator, including MCP server failures, GPU detection, and authentication errors. It provides a structured troubleshooting methodology for developers facing errors, performance problems, or system failures. Use it when you encounter operational issues, unexpected behavior, or need to isolate failing components.

The testing-assistant skill manages the complete testing lifecycle for the Claude Patent Creator, including unit, integration, and end-to-end validation. It provides expert guidance on running test suites, debugging failures, and maintaining proper test pyramid strategies. Developers should use it for validating features, testing after changes, and setting up CI/CD pipelines.

The setup-assistant skill guides developers through installing, configuring, and performing first-time setup for the Claude Patent Creator MCP server. It provides a complete setup lifecycle from pre-installation checks to verification, including quick setup scripts and troubleshooting steps. Use it for initial installation, environment configuration, authentication setup, or when migrating to a new machine.

This skill provides a systematic 7-step methodology for conducting comprehensive patent prior art searches and patentability assessments. It's designed for developers needing to search patent databases, analyze patent landscapes, or prepare for patent filings. Key capabilities include querying large patent datasets via BigQuery, performing CPC classification searches, and generating patentability reports.

This Claude skill enables patent searching across 76M+ patents via BigQuery for fast, zero-setup global searches, or through the PatentsView API for detailed US patent metadata. Developers can quickly test it using the provided Python script with pre-configured credentials. Choose BigQuery for worldwide coverage or PatentsView for rich US patent details like inventors and citations.

This skill provides automated review of utility patent applications against USPTO MPEP guidelines, enabling developers to check compliance and analyze claims, specifications, and formalities. It integrates with patent databases and USPTO regulations for comprehensive patent analysis and drafting assistance. Use it when building patent-related applications that require USPTO compliance checking or patent creation workflows.

This skill generates patent-style technical diagrams (flowcharts, block diagrams, system architectures) as SVG files using Graphviz. It automatically handles patent-style reference numbering and supports templates for common patterns. Use it when you need to create professional technical diagrams suitable for USPTO submissions or documentation.

This Claude skill generates patent-style technical diagrams like flowcharts and block diagrams using Graphviz, automatically adding reference numbering. It's designed for creating figures for method claims, system architectures, and other patent illustrations. Developers should invoke it when they need to quickly produce formal diagrams for patent documentation.

This skill analyzes patent claims for USPTO compliance, specifically checking for antecedent basis and definiteness under 35 USC 112(b). Use it to automatically review claim structure, identify drafting issues like missing term introductions, and validate claims before filing. It performs automated checks across claims to flag subjective language and improper references.

This skill automates the complete patent drafting workflow, from prior art search to generating a USPTO-ready filing package including claims, specifications, and diagrams. Use it when developers need to create a full provisional or utility patent application from an invention disclosure. It handles the key steps of drafting, compliance checking, and document generation end-to-end.

This skill provides hybrid RAG search across USPTO patent documentation (MPEP, statutes, regulations) with post-January 2024 updates. It combines FAISS vector search, BM25 keyword matching, HyDE, and cross-encoder reranking for precise results. Use it when you need to query or retrieve specific passages from official U.S. patent examination materials.