Meta-Prompt: Build Role-Specific Skills
A meta-prompt for drafting reusable skill files that encode role standards, operating rules, failure handling, and output expectations.
Use cases
Operations & Workflow, Strategy & Planning
Platforms
Claude, GPT, Model-Agnostic
Jump to a section
Workspace-ready resource
Copy the raw material, then check the setup, output contract, and failure modes.
Design a reusable AI skill for a specific role. Inputs to collect:- role name- audience or user- recurring task types- non-negotiable standards- forbidden behaviour- preferred output style- common failure cases Return:1. Skill title2. Skill description3. Full skill file in markdown4. Explanation of why each section exists5. What still needs human testing Requirements for the skill:- It must describe how the role operates, not just what tone it uses.- It must include a fallback for missing context.- It must stay narrow enough to debug later.Workspace translation
Turn this resource into an inspectable run.
Best next step
Run it once manually. If the same job appears twice, package it.
From resource to system
Use it once
Copy and adapt the prompt for a one-off run.
Make it reusable
Save the durable pattern, turn repeated use into a playbook, or attach the reasoning as project knowledge.
Decide after the run
Run it once manually. If the same job appears twice, package it.
Quality bar
Before using this, check the contract.
What input does this require?
What output should it produce?
Where can it fail?
What should a human review?
Recommended path
When to Use This
Use this when you want to turn a repeated working setup into a reusable skill instead of rewriting the same instructions every time.
It is useful for teams building libraries of role-specific AI behaviours: researcher, proposal writer, support triager, founder ghostwriter, and similar.
Why It Works
The prompt is designed to create operational roles, not character sketches. That distinction matters because good skills are about standards, fallback behaviour, and scope, not decorative adjectives.
The “what still needs human testing” section also matters. It stops the model from pretending a generated skill is already production-ready.
How to Customise
If you already have examples of good and bad outputs, provide them as part of the input. That makes the resulting skill much sharper.
You can also ask the meta-prompt to produce both a “light” and “strict” version of the same skill for different workflows.
Limitations
This creates first drafts, not finished capability packs. The generated skill still needs real testing across multiple tasks.
If the requested role is too broad, the meta-prompt will usually produce something vague. Narrow roles win.
Model Notes
Claude tends to produce cleaner rationale and better guardrails.
GPT is useful if you want the skill output in more rigid structure. Model-agnostic overall, but the quality of the input role definition sets the ceiling.
Related Resources
Browse PromptsMeta-Prompt: Generate Custom System Prompts
A prompt that generates system prompts. Describe what you need an AI to do, and this meta-prompt produces a structured, production-ready system prompt following best practices.
Pattern / playbook seed
Operations & Workflow · Strategy & Planning
Skill: Technical Documentation Writer
A skill file that configures an LLM to write clear, structured technical documentation. Handles API docs, setup guides, README files, and process documentation with consistent formatting and appropriate detail depth.
Operator seed
Development & Code · Operations & Workflow
Skill: Editorial Voice Configuration
A reusable skill file that gives an LLM a specific editorial voice. Defines tone, sentence structure, vocabulary rules, and anti-patterns. Drop it into any AI tool to maintain consistent brand voice across all content.
Operator seed
Content & Writing · Marketing & Growth
Related Guides
How to Turn a Good Prompt into a Reusable Skill
A practical guide to recognising when a one-off prompt has become a reusable capability and how to package it into a cleaner skill.
Custom Instructions vs Skills vs Agents
When to use light behavioural setup, reusable capability packs, or full multi-step systems without muddling the three together.
Need this operationalized?
Turn the pattern into a workspace system.
Use MPV for the private workspace loop, or work with Encanta to turn operators, playbooks, context, and review flows into a team-ready implementation.