Prompting7 min read
Few-Shot Prompting: When and How to Use Examples
What few-shot prompting is, when examples beat instructions, how many to use, how to pick and format them, and how to stop the model copying them too closely.
Fill in a few fields and get a well-structured prompt you can paste into any AI assistant. Start from a template or a blank page.
Start from a template
You are a careful analyst who writes clear briefings. ## Task Summarise the document below for a busy decision-maker. Highlight decisions, deadlines, risks and open questions. ## Tone Neutral and factual. ## Output format Bullet points, one idea per bullet. ## Length About 150–300 words. ## Rules - Only use information from the document - Mark anything unclear as an open question - Quote figures exactly as written - If you are not sure about something, say so clearly instead of guessing. If important information is missing, ask me up to three short clarifying questions before you start.
599 characters, about 150 tokens
Runs entirely in your browser. No data is sent to a server.
Any chat-based assistant or LLM API that accepts text instructions. The structure of role, task, context, format and rules is not tied to one vendor.
No. The prompt is assembled in your browser and never leaves your device unless you copy it somewhere yourself. Still avoid pasting confidential data into AI tools your company has not approved.
Delimiters make it clear where your source material starts and ends, so the model is less likely to confuse it with instructions.
No. A role helps when expertise, vocabulary or perspective matters. For simple tasks a clear task description and output format matter much more.
Prompting7 min read
What few-shot prompting is, when examples beat instructions, how many to use, how to pick and format them, and how to stop the model copying them too closely.
Prompting7 min read
How to write a system prompt for an AI assistant or automation: a proven structure, a full example for customer support, testing tips and the mistakes to avoid.
Prompting8 min read
Prompt engineering for business users: a six-part prompt structure, before-and-after examples, how to test prompts and how to share them in a team.
Prompting7 min read
Twelve ready-to-use prompt templates for business: emails, summaries, meeting notes, data extraction, reports and feedback, with tips to adapt them.
Estimate what an LLM API costs per request, month and year. Compare three price scenarios side by side and turn sample text into a token estimate.
Calculate the ROI of automating a task with AI: hours saved, net monthly benefit, payback period and first-year return, including review time and running costs.
Answer a few questions to see which EU AI Act risk category your AI system likely falls into and which obligations apply. Free orientation, not legal advice.