Saved prompts are dead. Agent Skills are the next step.
If you’ve been around for a while, you probably have a file full of bash one-liners.
Small scripts or commands you saved because they solved a problem you didn’t want to automate properly.
When coding agents arrived, prompts became the new one-liners.
Useful prompts were saved, reused, and eventually turned into “prompt files”, then slash commands like /do-something.
But that model has already evolved.
⚙️ Agent Skills
Agent Skills are the next iteration.
At a basic level, a skill looks a lot like a saved prompt: a directory with a markdown file.
What makes it different is how it’s used.
Skills include metadata like name and description, allowing agents to discover them.
Instead of explicitly calling a prompt every time, the agent can determine when to use a skill based on intent.
This is referred to as progressive disclosure:
- Agent loads skill metadata
- Matches it to your task
- Then loads and executes the full skill when needed
You can still call skills directly (/, $, @), but you don’t always have to.
🧠 More Than Just Prompts
The real differentiator is that skills aren’t just prompts.
They can include reference documentation, templates, and scripts.
This means you’re no longer just telling the agent what to do.
You’re giving it tools and context to execute and validate tasks.
For more complex workflows, it’s often easier to write a script and teach the agent how to use it than to encode everything in a prompt.
⚠️ A Word of Caution
This power comes with risk.
Skills can include executable logic and tell agents to perform tasks.
That means a shared skill can contain malicious or unsafe behavior.
Treat them like any script you install:
- Understand what they do
- Know where they come from
- Review before using (watch out for hidden text or obfuscated instructions)
🧠 Final Thoughts
Agent skills are a meaningful step forward.
They let you codify workflows, preferences, and repeatable agent tasks in a way that agents can discover.
They’re a strong productivity accelerator and a powerful way to capture institutional knowledge in a form agents can actually use.
(More on that in the next post.)