AI Agents Skills

This skill should be used when the user asks to "A/B test agent output", "compare AI-generated content", "test AI vs human output", "run a split test on agent emails", "compare agent variants", "test different prompts in production", "A/B test AI-written cold emails", "compare AI model outputs", "split test LLM-generated content", or any variation of A/B testing AI agent-generated content against other variants for B2B SaaS GTM.

This skill should be used when the user asks to "evaluate agent output", "measure agent quality", "build an eval for an AI agent", "assess LLM agent performance", "create evaluation criteria for agents", "test agent output quality", "build an agent eval framework", "score agent responses", "measure AI agent accuracy", or any variation of evaluating and measuring the quality of AI agent output for B2B SaaS GTM tasks.

This skill should be used when the user asks to "orchestrate multiple agents", "coordinate agents", "design a multi-agent system", "build an agent pipeline", "chain agents together", "hand off between agents", "design agent workflows", "build an agent DAG", "route between agents", or any variation of designing and implementing systems where multiple AI agents work together on GTM tasks for B2B SaaS.

This skill should be used when the user asks to "build an agent with the Claude SDK", "use the Claude Agent SDK", "create a Claude agent", "get started with the Anthropic agent SDK", "build an agentic app with Claude", "set up the Claude agent SDK", "write an agent with tool use", "build an agent in Python with Claude", "build an agent in TypeScript with Claude", or any variation of getting started with Anthropic's agent SDKs to build AI agents for B2B SaaS use cases.

This skill should be used when the user asks to "build a GTM agent", "design a sales agent", "create an outbound agent", "build an AI agent for go-to-market", "design an agent for prospecting", "build an agent for lead scoring", "create an AI SDR", "automate GTM with agents", "design a revenue agent", or any variation of designing AI agents that perform go-to-market tasks for B2B SaaS teams.

This skill should be used when the user asks to "design agent handoffs", "hand off between agents", "pass context between agents", "build multi-agent pipelines", "design agent-to-agent communication", "orchestrate agent handoffs", "structure agent pipelines", "pass state between AI agents", "design agent chains", or any variation of designing how AI agents hand off work to each other in B2B SaaS GTM workflows.

This skill should be used when the user asks to "write a system prompt", "design a prompt for an agent", "improve an agent prompt", "write instructions for an AI agent", "create a system prompt for a GTM agent", "fix my agent's prompt", "tune a prompt", "write a prompt for a research agent", "write a prompt for an email writer agent", or any variation of designing, writing, or improving system prompts for AI agents performing B2B SaaS go-to-market tasks.

This skill should be used when the user asks to "design tools for an AI agent", "build agent tools", "create function calling tools", "design tool schemas for LLMs", "build tools for Claude", "design agent capabilities", "write tool definitions", "create agent tool interfaces", "design function schemas for agents", or any variation of designing and building tools that AI agents call in B2B SaaS GTM workflows.