Skip to main content
Skills are pre-packaged sets of instructions, context, and references that you add to an AI assistant so it handles a specific task the way you want it handled—every time, without you having to re-explain. When you load a skill, the assistant gains a durable set of guardrails and know-how for a narrow job: how your team writes documentation, how to interpret your Conductor data, how to structure a weekly performance report, and so on. The assistant still answers you in natural language; the skill just makes those answers more consistent and more grounded in your context.

​
What a skill contains

You don’t need to build a skill in order to use one, but it helps to know what’s inside:
  • A short description. A plain-language summary of when the skill should be used. This helps the assistant decide, on its own, when to reach for it.
  • Instructions. The core of the skill—the prompts, rules, and examples that shape how the assistant should respond for this specific job.
  • Supporting files (optional). Reference documents, templates, style guides, or checklists the assistant can pull from when it needs them.
Most skills are packaged as a small folder or a .zip file. The exact format varies by platform, but the shape is the same across tools.

​
Why use a skill

Reach for a skill when:
  • You find yourself repeating the same instructions to your assistant (“always use our brand voice,” “always start with the AI search dataset”).
  • You want multiple teammates to get the same quality of output from the same AI tool.
  • You want answers that reflect your data and your terminology, not generic advice from the open web.
  • You want to hand work off to the assistant with confidence that it will follow a defined process.
Skills turn one-off prompt effort into a reusable asset your whole team can rely on.

​
How to use a skill in your LLM platform

The specifics differ by tool, but the pattern is the same everywhere:
1

Download the skill

Get the skill from the Conductor template gallery or another source you trust. You will usually receive a .zip file or a folder.
2

Install it into your assistant

Each platform has its own location for skills. Claude apps, ChatGPT, Cursor, Windsurf, and other tools each expose this a little differently. Your platform’s documentation will tell you exactly where to place the skill files or how to upload them.
3

Trigger the skill in conversation

Once installed, the assistant can apply the skill automatically when your question matches its purpose. You can also reference the skill by name if you want to be explicit (“use the Conductor School skill for this”).
4

Keep it updated

Skills evolve. When Conductor or another provider publishes a new version, replace the old files with the new ones so your assistant stays current.
Platform-specific setup guides live in the AI Tools section.

​
An example: Conductor School for LLMs

The Conductor School for LLMs skill teaches your assistant how to talk about Conductor—our product language, our datasets, how to read AI search performance data, and how to translate findings into recommendations a marketing team can actually use. On its own, the skill sharpens every Conductor-related answer the assistant gives. Paired with the Conductor MCP, it becomes far more useful:
  • The MCP gives the assistant live, authorized access to your Conductor data.
  • The skill tells the assistant how to interpret that data, which vocabulary to use, and which patterns lead to sound recommendations.
Together, you get answers that are grounded in your actual metrics and framed the way a Conductor expert would frame them—without having to coach the assistant every session.

​
Tips for getting the most out of skills

  • Start with one. Install a single skill, use it in a few real workflows, and get a feel for how the assistant behaves before layering more.
  • Combine skills with MCP where it makes sense. Skills shape how the assistant thinks. MCP gives it the data. The two are complementary, not overlapping.
  • Describe the job, not the tool. When you chat, focus on what you are trying to accomplish. A well-written skill will handle the rest.
  • Revisit regularly. As your team’s work evolves, swap in updated skills to keep your assistant aligned.