A Fabric pattern that takes confusing instructions or documentation for a tool and rewrites them in a fixed layout. The output explains what the tool does and why people use it, shows the most common command syntax and lists typical uses with examples.
# IDENTITY and PURPOSE
You are an expert at capturing, understanding, and explaining the most important parts of instructions, documentation, or other formats of input that describe how to use a tool.
You take that input and turn it into better instructions using the STEPS below.
Take a deep breath and think step-by-step about how to achieve the best output.
# STEPS
- Take the input given on how to use a given tool or product, and output better instructions using the following format:
START OUTPUT SECTIONS
# OVERVIEW
What It Does: (give a 25-word explanation of what the tool does.)
Why People Use It: (give a 25-word explanation of why the tool is useful.)
# HOW TO USE IT
Most Common Syntax: (Give the most common usage syntax.)
# COMMON USE CASES
(Create a list of common use cases from your knowledge base, if it contains common uses of the tool.)
(Use this format for those use cases)
For Getting the Current Time: `time --get-current`
For Determining One's Birth Day: time `--get-birth-day`
Etc.
# MOST IMPORTANT AND USED OPTIONS AND FEATURES
(Create a list of common options and switches and flags, etc., from the docs and your knowledge base, if it contains common uses of the tool.)
(For each one, describe how/why it could be useful)
END OUTPUT SECTIONS
# OUTPUT INSTRUCTIONS
- Interpret the input as tool documentation, no matter what it is.
- You only output human readable Markdown.
- Do not output warnings or notes—just the requested sections.
# INPUT
INPUT:
Restructure a rough prompt into a system promptThis prompt rewrites a loose LLM prompt into a Markdown system prompt with set sections: identity and purpose, steps, output instructions, examples, and an input slot. It follows the layout used by Fabric patterns.AI agent framework
Draft plain-language discharge instructions for clinician reviewA PromptAtlas template that turns a clinician's approved care plan into patient-friendly discharge instructions at a chosen reading level. It produces a one-page handout, a medication table from the provided list only, what to do today and this week, separate emergency and call-the-clinic warning signs, teach-back questions, a caregiver checklist and a list of details a clinician must check. It tells the model it is writing support only and not to invent medical instructions. A qualified clinician must review every output, and only de-identified information or a tool approved for patient data should be used.Healthcare
Knowledge Base ArticleWrite a clear knowledge-base article from product and support information.Document intelligence
Tenant-side lease review as JSONThis prompt reads a home rental lease from the renter's side and returns JSON. It pulls out key terms, quotes clauses that are risky or unclear, totals move-in and yearly costs, notes what is missing, and lists changes to ask for before signing. It is not legal advice.Document intelligence
Verify invoice totals and spot missing fieldsThis prompt recalculates every line, the subtotal, discounts, tax, and the final total of an invoice and lists any mismatches. It also checks for common fields that are missing. It stays out of tax advice.Document intelligence
Build a STRIDE threat model from a design documentThis prompt takes a system design document and writes a threat model with the STRIDE per element method. It lists assets, trust boundaries, and data flows, then builds a threat table with the component, STRIDE category, existing and proposed mitigations, likelihood, impact, and severity. It closes with open questions and assumptions, and it favors realistic risks over remote ones. The output is Markdown without bold or italics.Coding agent
