Prompts: Reusable Instructions
Templated messages users can trigger on demand.
Prompts: Reusable Instructions is a free MCP Academy lesson on CoddyKit — lesson 3 of 4. You can read the complete lesson below for free — then practise it hands-on in the browser with a built-in code editor and a 24/7 AI tutor. It is part of the MCP Academy learning path, one of 4 lessons in the course, and your progress syncs across the web and the CoddyKit app.
The Third Primitive
The last building block is the prompt: a reusable, templated message a user can trigger to start a task the right way. 💬
Prompts Are Templates
Think of a prompt as a saved instruction with blanks to fill in. Instead of retyping a long request, the user picks it and supplies a few inputs.
User-Controlled by Design
Prompts are user-controlled: the person explicitly chooses one, often from a slash menu or button. The model does not summon them on its own.
Declaring a Prompt
In Python you mark a function with @mcp.prompt(). It returns the text that becomes the starting message for the conversation.
@mcp.prompt()
def summarize(text: str) -> str:
return f"Summarize this:\n{text}"Arguments Fill the Blanks
A prompt's function parameters become its fillable slots. The host shows fields, the user types values, and they slot into the template.
Output Can Be Messages
A prompt can return more than one line. It may produce a list of messages with roles, shaping a richer multi-turn starting point.
Discoverable Like the Rest
Clients can list the prompts a server offers, just as they list tools and resources. That is how a slash-command menu gets populated.
Consistency for Everyone
A shared prompt captures a proven way to ask. Everyone on a team gets the same high-quality phrasing without memorizing it.
A Code-Review Example
A review prompt might take a code snippet and frame it: please review this for bugs and style. The user just pastes the code.
@mcp.prompt()
def review(code: str) -> str:
return f"Review this code for bugs:\n{code}"Not a Tool, Not a Resource
A prompt does not run code or fetch data. It simply seeds the conversation with well crafted instructions for the model to follow.
Where Prompts Shine
Reach for a prompt for repeatable workflows: summarize this, explain this error, draft a release note. Common asks become one click.
Quick Check
Let's pin down who controls a prompt.
Recap: Prompts
You've got it! A prompt is a user-triggered template with fillable arguments. It returns ready-made messages so common requests stay consistent. ✅
Frequently asked questions
Is the “Prompts: Reusable Instructions” lesson free?
Yes — the full text of “Prompts: Reusable Instructions” is free to read here on the web, and the MCP Academy course includes 4 lessons in total. To practise it interactively (a built-in code editor and a 24/7 AI tutor) and unlock the rest of the MCP Academy course, upgrade to CoddyKit PRO.
What will I learn in “Prompts: Reusable Instructions”?
Templated messages users can trigger on demand. You practise MCP Academy with hands-on code you run directly in the browser, and a 24/7 AI tutor answers your questions as you work through the lesson.
Do I need any experience to start MCP Academy?
No prior experience is required. MCP Academy on CoddyKit is structured for beginners through advanced learners; this is — lesson 3 of 4, so you can start here or from the beginning and move at your own pace.
How long does the “Prompts: Reusable Instructions” lesson take?
Most CoddyKit lessons take about 5–10 minutes. Each one is bite-sized and interactive, so you make steady progress and pick up exactly where you left off across the web and the app.
Can I write and run code in this MCP Academy lesson?
Yes. Every MCP Academy lesson includes a built-in code editor, so you write and run real code right in your browser and get instant AI feedback — no local setup required.
All lessons in this course
- Tools: Actions the Model Takes
- Resources: Data the Model Reads
- Prompts: Reusable Instructions
- Picking the Right Primitive