---
title: "Prompt templates for the work that travels"
description: "Eighteen copy-ready prompts for client work, sales, research, writing, admin, coding, agent operations and learning — each with inputs, an output contract and the checks before you trust it."
canonical: "https://www.diginomads.win/prompts"
source: "content/pages/prompts/index.md"
updatedAt: "2026-11-10"
---

## Cards

- [Scope a client project from a messy call](/prompts/scope-a-project-from-a-call): Turn a rambling discovery call into a one-page scope: outcomes, deliverables, exclusions, assumptions, dependencies and open questions. (8 min · Any chat model)
- [Write a weekly update that prevents scope creep](/prompts/weekly-client-update): A status update that separates done, next, blocked and decisions needed — and makes any change to the plan visible before it becomes unpaid work. (6 min · Any chat model)
- [Qualify a lead before you write a proposal](/prompts/qualify-a-lead): Extract the problem, urgency, budget signal and decision process, then decide whether to pursue, nurture or decline before spending a day on a proposal. (7 min · Any chat model)
- [Follow up after a proposal without chasing](/prompts/follow-up-after-proposal): A short follow-up that names the decision, offers three next steps and leaves room for a no — without discounts or fake urgency. (6 min · Any chat model)
- [Build a city brief from official sources](/prompts/city-research-brief): Price a city from statistics offices, fee schedules and tax authorities, with a cost floor, a newcomer premium and a confidence column. (10 min · Model with web access)
- [Compare visa routes in a decision matrix](/prompts/visa-route-matrix): Turn marketing pages into a criteria matrix: what the permit grants, who may hold it, which income counts and which questions need a professional. (12 min · Model with web access)
- [Rewrite a draft for one named reader](/prompts/rewrite-for-a-reader): Cut the throat-clearing, keep the claim and the caveat, and produce a version a specific person can act on without a second read. (7 min · Any chat model)
- [Turn notes into a publishable article](/prompts/notes-to-article): Extract claims, separate fact from opinion, build an outline with evidence slots, then draft without inventing a source or a number. (12 min · Any chat model)
- [Turn receipts into a bookkeeping-ready log](/prompts/receipt-log): Normalise dates, merchants, currencies and tax notes into a CSV, with an explicit unknown column and a duplicate check instead of guessed data. (9 min · Any chat model)
- [Build a travel admin checklist from a booking](/prompts/booking-admin-checklist): Extract every deadline, document, dependency and buffer from a booking confirmation, then assign an owner and a latest-start date. (6 min · Any chat model)
- [Reproduce a bug before fixing it](/prompts/reproduce-a-bug): Make the agent write the failing test first, isolate the smallest cause and only then patch — with the command that proves it. (10 min · Coding harness)
- [Review a diff for operational risk](/prompts/review-a-diff): Review changed behaviour, failure modes, rollback, secrets and tests. Findings only, ordered by severity, with file and line references. (8 min · Coding harness)
- [Write the project brief your agent actually reads](/prompts/write-agents-md): Turn the commands, boundaries and review path of a repository into a short project instruction file an agent can follow without a preamble. (8 min · Any coding harness)
- [Audit an agent workflow before you automate it](/prompts/audit-an-agent-workflow): Map a workflow step by step, separate drafting from deciding, and put human checkpoints where a wrong output is expensive and hard to detect. (9 min · Any chat model)
- [Turn a repeated prompt into a skill](/prompts/turn-a-prompt-into-a-skill): Convert the prompt you paste every week into an Agent Skills folder with a specific trigger, an output contract and two pass tests. (8 min · Any chat model)
- [Design a fallback path for a flaky tool](/prompts/design-a-fallback-path): Classify how a dependency fails, decide what retries and what escalates, and write the runbook before the tool goes down at the wrong moment. (9 min · Any chat model)
- [Learn a tool by building a tiny artifact](/prompts/learn-a-tool-by-building): A five-session curriculum where each session ends in a small working thing, a test of understanding and one source to read — no tutorial collecting. (10 min · Any chat model)
- [Red-team a plan before you commit](/prompts/red-team-a-plan): List the assumptions, the failure modes and the cheapest test that would change your mind, then write the kill criteria before you start. (9 min · Any chat model)

## What is in every template

Each page follows the same shape, so you can scan one and know where to look on the next:

- **Use when** — the situation the prompt is built for, and the one it is not.
- **Inputs you need** — the evidence to paste in. Verbatim beats a summary.
- **Copy-ready prompt** — the full text, variables in square brackets. No template token, no hidden instruction.
- **Output contract** — what a good answer contains, so a missing section is obvious.
- **Guardrails** — the checks that stop a confident answer from becoming a wrong decision.
- **Variations** — the one edit to make when the first run is close but not right.

## How to adapt a template

Replace the bracketed variables first. Then change one instruction at a time, run the same input, and keep the version that produces the better output. If you change three instructions at once, you learn nothing about which one helped.

If you use the same prompt more than twice a week, stop pasting it: turn it into a [skill](/skills) with a trigger and a pass condition. If the prompt depends on files, commands or a repository, the [harness comparison](/harnesses) is the next decision.

## What these prompts will not do

They will not make a legal, tax, immigration or financial decision for you. They will not turn a bad source into a good one. They will not know your client's internal politics, your passport or your risk tolerance. Use them to structure the work and surface the questions; keep the judgement.

## For agents

Every prompt is inside its page body and in the Markdown mirror at `/prompts/<slug>.md`. The collection itself is at `/prompts.md`. If a template assumes a workflow that does not exist, submit the gap through the [help desk](/agent-help) with the scenario, the expected output and the output you got.

## Structured data

```json
{
  "@context": "https://schema.org",
  "@type": "WebPage",
  "name": "Prompt templates for the work that travels",
  "url": "https://www.diginomads.win/prompts",
  "description": "Eighteen copy-ready prompts for client work, sales, research, writing, admin, coding, agent operations and learning — each with inputs, an output contract and the checks before you trust it.",
  "isPartOf": {
    "@type": "WebSite",
    "name": "DigiNomads",
    "url": "https://www.diginomads.win/"
  }
}
```