skills
(636)agent-evaluation
Evaluate LLM agents using behavioral testing, capability assessment, reliability metrics, and production monitoring to identify issues before deployment. This command helps bridge the gap between benchmark performance and real-world reliability by addressing the unique challenges of agent evaluation.
npx claude-code-templates@latest --skill=ai-research/agent-evaluation --yesagent-manager-skill
Orchestrate and schedule multiple local CLI agents using tmux sessions, enabling parallel execution, monitoring, and task assignment.
npx claude-code-templates@latest --skill=ai-research/agent-manager-skill --yesagent-memory-mcp
Provides a persistent, searchable memory bank for AI agents, enabling them to record, retrieve, and analyze knowledge and decisions related to a project. It includes MCP tools for searching, writing, reading, and viewing statistics on memory usage, along with a dashboard for visualization.
npx claude-code-templates@latest --skill=ai-research/agent-memory-mcp --yesagent-memory-systems
Explore the architecture of intelligent agent memory, including short-term context, long-term vector stores, and cognitive architectures that optimize memory retrieval. This skill focuses on effective chunking, embedding, and retrieval strategies to ensure agents remember and utilize information effectively.
npx claude-code-templates@latest --skill=ai-research/agent-memory-systems --yesagent-tool-builder
This skill helps you design effective tools for AI agents by focusing on clear schema definitions, comprehensive descriptions, and robust error handling. It emphasizes that well-documented tool schemas are more crucial than complex code implementations for successful agent interactions.
npx claude-code-templates@latest --skill=ai-research/agent-tool-builder --yesagents-autogpt
Build and deploy autonomous AI agents and complex automation systems using a visual workflow builder and comprehensive development toolkit. AutoGPT enables the creation of continuous, triggered agents with modular components and performance benchmarking.
npx claude-code-templates@latest --skill=ai-research/agents-autogpt --yesagents-crewai
Orchestrate autonomous AI agents in collaborative teams to solve complex tasks using role-based delegation, memory, and sequential/hierarchical process execution, all without LangChain dependencies [cite: null]. This framework enables building production-ready multi-agent systems with specialized agents and over 50 built-in tools [cite: null].
npx claude-code-templates@latest --skill=ai-research/agents-crewai --yesagents-langchain
Build LLM-powered applications using LangChain, a framework that simplifies agents, chains, and RAG implementation with extensive integrations and provider support. This tool enables rapid prototyping and production deployment of chatbots, question-answering systems, and autonomous agents.
npx claude-code-templates@latest --skill=ai-research/agents-langchain --yesagents-llamaindex
Build LLM applications with retrieval-augmented generation (RAG) using LlamaIndex, a data framework specializing in document ingestion, indexing, and querying with 300+ connectors. This tool is ideal for document Q&A, chatbots, knowledge retrieval, and constructing RAG pipelines for data-centric LLM applications.
npx claude-code-templates@latest --skill=ai-research/agents-llamaindex --yesai-agents-architect
Architect and construct autonomous AI agents with robust tool use, memory management, and planning, while designing for controllable behavior and graceful failure. This skill specializes in building AI agents, emphasizing tool and function calling, and multi-agent orchestration.
npx claude-code-templates@latest --skill=ai-research/ai-agents-architect --yesautonomous-agent-patterns
Explore design patterns for building autonomous coding agents, covering core agent architectures, tool integration, permission systems, browser automation, context management, and human-in-the-loop workflows. Use these patterns to build AI agents with tool/function calling APIs, permission/approval systems, and sandboxed execution.
npx claude-code-templates@latest --skill=ai-research/autonomous-agent-patterns --yesautonomous-agents
Explore the principles of designing reliable autonomous AI systems that can decompose goals, plan actions, and self-correct, emphasizing the importance of constrained autonomy and error mitigation for production-ready agents. This skill covers agent loops, goal decomposition, reflection patterns, and production reliability.
npx claude-code-templates@latest --skill=ai-research/autonomous-agents --yesbehavioral-modes
Adapts the AI's behavior and communication style to optimize for specific tasks like brainstorming, coding, debugging, reviewing, teaching, and shipping. This skill helps the AI respond more effectively by automatically detecting or manually switching between different operational modes.
npx claude-code-templates@latest --skill=ai-research/behavioral-modes --yesclaude-code-guide
Provides a comprehensive guide to effectively use Claude Code, covering configuration, prompting strategies, debugging, and best practices. It leverages a `CLAUDE.md` file for project-specific guidance and advanced prompting techniques.
npx claude-code-templates@latest --skill=ai-research/claude-code-guide --yescomputer-use-agents
Build AI agents that can interact with computer interfaces like humans, using vision to perceive the screen and executing actions via mouse and keyboard. This command provides patterns and implementations for developing and deploying such agents, with a strong emphasis on sandboxing and security.
npx claude-code-templates@latest --skill=ai-research/computer-use-agents --yescontext-window-management
Optimize large language model (LLM) context windows by employing strategies like summarization, trimming, and intelligent routing to overcome token limits and context degradation, ensuring focused and relevant information processing.
npx claude-code-templates@latest --skill=ai-research/context-window-management --yescontext7-auto-research
Automatically fetches the latest library and framework documentation for Claude Code using the Context7 API, providing up-to-date information for libraries like React, Next.js, and Prisma.
npx claude-code-templates@latest --skill=ai-research/context7-auto-research --yesconversation-memory
Implement persistent memory for LLM conversations, utilizing short-term, long-term, and entity-based memory to enhance context and relevance across interactions. It manages memory lifecycles, ensures intelligent retrieval, and maintains user isolation.
npx claude-code-templates@latest --skill=ai-research/conversation-memory --yescrewai
Orchestrates AI agent teams using the CrewAI framework, allowing you to define agent roles, tasks, and workflows for complex problem-solving. This command helps you design and manage collaborative AI agents, leveraging sequential, hierarchical, or parallel processes for optimal team performance.
npx claude-code-templates@latest --skill=ai-research/crewai --yesdata-processing-nemo-curator
Use data-processing-nemo-curator to leverage GPU-accelerated tools for cleaning, deduplicating, and curating multi-modal datasets (text, image, video, audio) for LLM training, featuring fuzzy deduplication, quality filtering, PII redaction, and NSFW detection. Scale data processing across GPUs with RAPIDS to prepare high-quality training datasets or clean web data.
npx claude-code-templates@latest --skill=ai-research/data-processing-nemo-curator --yes