Beyond Prompting: Clare Kitching on Teaching AI Your Unique ‘Skills’

C

Clare Kitching

LinkedIn Author

Transform your AI & data ambition into action | xQuantumBlack, xMcKinsey | Global top 100 Innovators in Data & Analytics | AI & data strategy, governance and capability building

In a recent LinkedIn post, Clare Kitching discusses the evolution of interacting with artificial intelligence, moving beyond basic prompting to the development of reusable ‘skills’ that teach AI a specific user’s method. Kitching argues that the current focus on improving prompts is only a stepping stone, and the next significant advancement lies in instructing AI on how individuals actually perform their tasks.

Kitching highlights the core concept of a skill in this context:

“A skill is a reusable set of instructions that shows AI your method.”

This approach, according to Kitching, allows users to bypass repetitive explanations for recurring tasks. Instead of re-explaining how to perform a specific action each time, users can impart their unique workflow to the AI.

The Power of Reusable Instructions

Clare Kitching elaborates on the practical implications of this skill-based teaching. She points out that users can instruct AI on various aspects of their work, such as:

  • How to review a proposal
  • How to transform meeting notes into an executive summary
  • How to research a potential client or prospect
  • How to structure the final output
  • What checks are necessary before completion

As Kitching explains, once these skills are established, an AI agent can access and apply them automatically whenever a similar task arises. This eliminates the need for constant re-instruction, thereby increasing efficiency.

Democratizing AI Skill Creation

A key point Kitching emphasizes is that creating these skills does not require advanced technical expertise. She states:

“And you don’t need to be technical to create one.”

According to Clare Kitching, if an individual understands how a particular task is executed, they can articulate that process. By explaining the workflow and providing examples, they can then leverage AI to help codify this into a reusable skill. This democratizes the ability to customize AI for specific, nuanced workflows.

Prompting vs. Skills

Kitching draws a clear distinction between the current state of prompting and the future potential of skills. She succinctly defines the difference:

“Prompting = asking AI to do something. Skills = teaching AI how you do it.”

This reframing suggests that true AI integration for productivity involves a shift from requesting actions to imparting methodologies. Kitching encourages users to begin by identifying a single task they perform weekly and work on developing a skill around it.

Clare Kitching’s insights suggest a more sophisticated and personalized way to harness AI, moving beyond generic commands to embedding unique human expertise directly into AI workflows.

📝 About This Content

This article is based on insights shared by Clare Kitching on LinkedIn.

📅 Originally posted on September 1, 2026 | View original post on LinkedIn →