{"course":{"slug":"prompting","title":"Getting Real Work out of a Model","url":"https://addaly.com/learn/prompting","level":"beginner","track":"start"},"lesson":{"slug":"handing-it-to-someone-else","title":"A prompt somebody else can run","url":"https://addaly.com/learn/prompting/handing-it-to-someone-else","position":72,"of":72,"minutes":9,"answer":"A prompt transfers only with its context, its expected output, and its \"when to distrust it\" section, and if you cannot write that section, you do not yet know when to distrust it yourself.","keyIdea":"A prompt transfers only with its context, its expected output, and its \"when to distrust it\" section, and if you cannot write that section, you do not yet know when to distrust it yourself."},"body":{"format":"markdown","text":"## The last step, and the one that proves you understood\n\nYou have a prompt that works. You give it to a colleague. It works less well for them, and neither of you can say why.\n\nMaking a prompt transferable is the final skill in this course, because it forces every implicit thing into the open, and the implicit things are where the quality was.\n\n## The four things that do not travel\n\n**Your context.** The prompt worked because you always paste the same kind of invoice, from the same supplier, in the same format. Your colleague pastes something else.\n\n**Your standing brief and memory.** Half your quality may be coming from custom instructions your colleague does not have. This is the commonest cause of a prompt that \"does not work for me\", and it is invisible from both sides.\n\n**Your judgement about the output.** You know that a `NOT_STATED` in the VAT column means go and look at the paper copy. Your colleague sees a completed table.\n\n**Your product setup.** Your tool has search and file reading; theirs is a different tier or a different product entirely.\n\n## What to hand over\n\nNot a prompt. A short document:\n\n```\n## Supplier quote comparison\nFor: comparing two or more quotes on the same criteria.\nTested: 2026-09-04, [model], works.\nNeeds: nothing special — plain chat, no search required.\n\nPROMPT\n[the prompt, with slots in CAPITALS]\n\nWHAT TO PASTE\nBoth quotes in full, including the terms page. Not summaries.\n\nWHAT GOOD LOOKS LIKE\nOne row per supplier. Every figure has a quoted source line.\nMissing figures say \"not stated\" — that is correct, not a failure.\n\nWHEN TO DISTRUST IT\n- If any row has no source quote, check that figure by hand.\n- If the two quotes are on different tax bases, the comparison is\n  invalid; check this first.\n- Lead times are invented if the \"do not estimate\" line is removed.\n  Do not remove it.\n\nIF IT GOES WRONG\nCheck the quotes were pasted in full. Then check whether your\ncustom instructions are adding anything.\n```\n\nSix headings, ten minutes to write, and it survives you leaving the organisation.\n\nThe **when to distrust it** section is the one nobody writes and the one that matters most. A prompt handed over without it is a machine with no dial markings.\n\n## Test it on a real person\n\nGive it to a colleague and watch them use it once, without helping. You will learn more in five minutes than from any amount of re-reading.\n\nWhat you will see: they paste a summary instead of the document. They stop at the first output rather than reading the source column. They remove the odd-looking line because it looks like clutter. Each of those is a fix to the document, not to them.\n\n## The team version\n\nThree files, in a place everybody can reach: the prompts, the examples, and the test items. Plain text, not a system that needs maintaining.\n\nTwo rules keep it alive: **every prompt has an owner**, and **every prompt has a date**. An unowned prompt library becomes a graveyard of things nobody trusts within about six months, and the failure is silent: people stop using it rather than reporting that it broke.\n\n## What this actually teaches you\n\nWriting the handover forces you to say what the prompt assumes, where it fails, and how you would know. Those are exactly the things you were carrying in your head and calling skill.\n\nIf you cannot write the \"when to distrust it\" section, you do not yet know when to distrust it, which is worth finding out while it is still your own work rather than somebody else's.\n\n## The end of the course, in one paragraph\n\nYou have three files: prompts that worked, examples from your own archive, and ten test items. You know that the model receives one flat block of text and nothing else, that agreement is its cheapest output, that fluency is not evidence, and that the checking has to be designed before the prompt is sent rather than performed after the answer arrives. Everything else here was an elaboration of those, and the files are what turn them into a practice rather than a set of things you once read."},"check":{"question":"A colleague reports that a shared prompt gives him worse results than its author gets. What should be checked first?","options":[{"text":"Whether he is using the same model version.","correct":false,"tempting":"Worth checking, and less likely than the difference that is invisible to both of them."},{"text":"Whether he pasted the prompt exactly, without typos.","correct":false,"tempting":"Small wording differences rarely produce a consistent quality gap."},{"text":"Whether the task is one he understands well enough to judge the output.","correct":false,"tempting":"A reason he might misjudge quality, not a reason the output would actually be worse."},{"text":"Whether the author's custom instructions, memory or project settings are supplying context the prompt never contained.","correct":true}]},"prev":{"slug":"applying-for-a-job","title":"Applications, without the generic letter","url":"https://addaly.com/learn/prompting/applying-for-a-job"},"next":null,"authorship":"Founder and series editor: Dr Mustafa Mun. Written by AI models to a written standard. Exam keys checked blind by a second model: 3,665 of 3,666 agreed. Academic review invited.","review":{"status":"Draft, open for academic review"},"cite":"Addaly (2026) 'A prompt somebody else can run', in Getting Real Work out of a Model. Edited by M. Mun. Addaly. Available at: https://addaly.com/learn/prompting/handing-it-to-someone-else (Accessed: 13 October 2026).","licence":{"name":"CC BY-NC-SA 4.0","url":"https://creativecommons.org/licenses/by-nc-sa/4.0/"},"reader":{"schema":1,"title":"A prompt somebody else can run","courseTitle":"Getting Real Work out of a Model","module":{"index":8,"slug":"the-work-you-actually-have","title":"The work you actually have"},"answer":null,"keyIdea":"A prompt transfers only with its context, its expected output, and its “when to distrust it” section, and if you cannot write that section, you do not yet know when to distrust it yourself.","hero":{"kind":"plate","image":null,"alt":null,"caption":null},"blocks":[{"t":"h2","id":"the-last-step-and-the-one-that-proves-you-understood","runs":[{"text":"The last step, and the one that proves you understood"}]},{"t":"p","runs":[{"text":"You have a prompt that works. You give it to a colleague. It works less well for them, and neither of you can say why."}]},{"t":"p","runs":[{"text":"Making a prompt transferable is the final skill in this course, because it forces every implicit thing into the open, and the implicit things are where the quality was."}]},{"t":"h2","id":"the-four-things-that-do-not-travel","runs":[{"text":"The four things that do not travel"}]},{"t":"p","runs":[{"text":"Your context.","b":true},{"text":" The prompt worked because you always paste the same kind of invoice, from the same supplier, in the same format. Your colleague pastes something else."}]},{"t":"p","runs":[{"text":"Your standing brief and memory.","b":true},{"text":" Half your quality may be coming from custom instructions your colleague does not have. This is the commonest cause of a prompt that “does not work for me”, and it is invisible from both sides."}]},{"t":"p","runs":[{"text":"Your judgement about the output.","b":true},{"text":" You know that a "},{"text":"NOT_STATED","code":true},{"text":" in the VAT column means go and look at the paper copy. Your colleague sees a completed table."}]},{"t":"p","runs":[{"text":"Your product setup.","b":true},{"text":" Your tool has search and file reading; theirs is a different tier or a different product entirely."}]},{"t":"h2","id":"what-to-hand-over","runs":[{"text":"What to hand over"}]},{"t":"p","runs":[{"text":"Not a prompt. A short document:"}]},{"t":"code","lang":"","label":"Text","text":"## Supplier quote comparison\nFor: comparing two or more quotes on the same criteria.\nTested: 2026-09-04, [model], works.\nNeeds: nothing special — plain chat, no search required.\n\nPROMPT\n[the prompt, with slots in CAPITALS]\n\nWHAT TO PASTE\nBoth quotes in full, including the terms page. Not summaries.\n\nWHAT GOOD LOOKS LIKE\nOne row per supplier. Every figure has a quoted source line.\nMissing figures say \"not stated\" — that is correct, not a failure.\n\nWHEN TO DISTRUST IT\n- If any row has no source quote, check that figure by hand.\n- If the two quotes are on different tax bases, the comparison is\n  invalid; check this first.\n- Lead times are invented if the \"do not estimate\" line is removed.\n  Do not remove it.\n\nIF IT GOES WRONG\nCheck the quotes were pasted in full. Then check whether your\ncustom instructions are adding anything."},{"t":"p","runs":[{"text":"Six headings, ten minutes to write, and it survives you leaving the organisation."}]},{"t":"p","runs":[{"text":"The "},{"text":"when to distrust it","b":true},{"text":" section is the one nobody writes and the one that matters most. A prompt handed over without it is a machine with no dial markings."}]},{"t":"h2","id":"test-it-on-a-real-person","runs":[{"text":"Test it on a real person"}]},{"t":"p","runs":[{"text":"Give it to a colleague and watch them use it once, without helping. You will learn more in five minutes than from any amount of re-reading."}]},{"t":"p","runs":[{"text":"What you will see: they paste a summary instead of the document. They stop at the first output rather than reading the source column. They remove the odd-looking line because it looks like clutter. Each of those is a fix to the document, not to them."}]},{"t":"h2","id":"the-team-version","runs":[{"text":"The team version"}]},{"t":"p","runs":[{"text":"Three files, in a place everybody can reach: the prompts, the examples, and the test items. Plain text, not a system that needs maintaining."}]},{"t":"p","runs":[{"text":"Two rules keep it alive: "},{"text":"every prompt has an owner","b":true},{"text":", and "},{"text":"every prompt has a date","b":true},{"text":". An unowned prompt library becomes a graveyard of things nobody trusts within about six months, and the failure is silent: people stop using it rather than reporting that it broke."}]},{"t":"h2","id":"what-this-actually-teaches-you","runs":[{"text":"What this actually teaches you"}]},{"t":"p","runs":[{"text":"Writing the handover forces you to say what the prompt assumes, where it fails, and how you would know. Those are exactly the things you were carrying in your head and calling skill."}]},{"t":"p","runs":[{"text":"If you cannot write the “when to distrust it” section, you do not yet know when to distrust it, which is worth finding out while it is still your own work rather than somebody else’s."}]},{"t":"h2","id":"the-end-of-the-course-in-one-paragraph","runs":[{"text":"The end of the course, in one paragraph"}]},{"t":"p","runs":[{"text":"You have three files: prompts that worked, examples from your own archive, and ten test items. You know that the model receives one flat block of text and nothing else, that agreement is its cheapest output, that fluency is not evidence, and that the checking has to be designed before the prompt is sent rather than performed after the answer arrives. Everything else here was an elaboration of those, and the files are what turn them into a practice rather than a set of things you once read."}]}],"check":{"question":[{"text":"A colleague reports that a shared prompt gives him worse results than its author gets. What should be checked first?"}],"options":[{"text":[{"text":"Whether he is using the same model version."}],"correct":false,"tempting":[{"text":"Worth checking, and less likely than the difference that is invisible to both of them."}]},{"text":[{"text":"Whether he pasted the prompt exactly, without typos."}],"correct":false,"tempting":[{"text":"Small wording differences rarely produce a consistent quality gap."}]},{"text":[{"text":"Whether the task is one he understands well enough to judge the output."}],"correct":false,"tempting":[{"text":"A reason he might misjudge quality, not a reason the output would actually be worse."}]},{"text":[{"text":"Whether the author’s custom instructions, memory or project settings are supplying context the prompt never contained."}],"correct":true,"tempting":null}],"order":[1,0,2,3],"why":null},"prev":{"slug":"applying-for-a-job","title":"Applications, without the generic letter","minutes":9},"next":null}}