[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78AQ6EEB32371FWMQP73KBG":3,"skill-same:01M78AQ6EEB32371FWMQP73KBG":64},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":35,"commitSha":36,"skillDirectory":37,"forksCount":38,"repositoryHtmlUrl":39,"repositoryHomepageUrl":40,"ownerHtmlUrl":41,"ownerBlogUrl":6,"ownerDescription":42,"updatedAt":43,"install":44,"resources":57},"01M78AQ6EEB32371FWMQP73KBG","task-brief",null,"Use when the user says \"brief this task\", \"scope this properly\", \"what do you need from me\", \"improve my prompt\", \"help me get better output\", \"turn this into a reusable prompt\", \"a prompt I can hand to a colleague\", or hands over a SUBSTANTIAL task — multi-step, multi-document, or a deliverable someone else will read — as a thin one-line request. Compile a task brief BEFORE starting substantial work — pin down the goal with checkable \"done when\" tests, context, constraints (standing rules included automatically), and tooling — confirm with the user, then execute against it. Reusable-prompt mode saves the brief as a structured prompt for recurring work instead of running it. Asks at most 2-3 clarifying questions, only where the answer would change the output; otherwise states assumptions in the brief and proceeds. Not for quick one-step asks — answer those directly.","moonlight-lupin\u002Fagent-skills","moonlight-lupin","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F136997773?v=4","01M78AEDSH1N62R0RNWPWMWVQY","01M78AEDT8ZF85MTK6XYBQBTNT",123,"MIT","https:\u002F\u002Fgithub.com\u002Fmoonlight-lupin\u002Fagent-skills\u002Fblob\u002Fmain\u002Fproductivity\u002Ftask-brief\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: task-brief\ndescription: \"Use when the user says \\\"brief this task\\\", \\\"scope this properly\\\", \\\"what do you need from me\\\", \\\"improve my prompt\\\", \\\"help me get better output\\\", \\\"turn this into a reusable prompt\\\", \\\"a prompt I can hand to a colleague\\\", or hands over a SUBSTANTIAL task — multi-step, multi-document, or a deliverable someone else will read — as a thin one-line request. Compile a task brief BEFORE starting substantial work — pin down the goal with checkable \\\"done when\\\" tests, context, constraints (standing rules included automatically), and tooling — confirm with the user, then execute against it. Reusable-prompt mode saves the brief as a structured prompt for recurring work instead of running it. Asks at most 2-3 clarifying questions, only where the answer would change the output; otherwise states assumptions in the brief and proceeds. Not for quick one-step asks — answer those directly.\"\nlicense: MIT\nmetadata:\n  version: 1.3.0\n  author: moonlight-lupin\n  platforms: [linux, macos, windows]\n  tags: [planning, brief, scoping, productivity, prompts]\n---\n\n# Task Brief\n\nThin requests are the biggest cause of mediocre agent output: the agent guesses the\ndeliverable, the audience and the guardrails, and the user only discovers the\nmisreading after the work is done. This skill moves that discovery to the **cheapest\npossible point** — before the work starts — by having the agent compile a short\n**task brief**, show it, and only then execute. The user never has to write a better\nprompt; the agent does the structuring.\n\nIt has two modes:\n- **Brief and run** (default) — compile the brief, confirm, then do the work.\n- **Reusable prompt** — compile the same brief, but save it as a structured prompt\n  for a task that recurs (a monthly report, a scheduled job, a prompt handed to a\n  colleague) instead of running it. See [Reusable-prompt mode](#reusable-prompt-mode).\n\n## When to run it — and when not to\n\n**Run it when:**\n- the user explicitly asks (\"brief this first\", \"make sure you understand before starting\"); or\n- the task is **substantial** — multi-step, multi-document, or a deliverable someone\n  else will read (a memo, a report, a dashboard, a document batch) — and the request\n  is thin or ambiguous enough that two reasonable readings would produce different work.\n\n**Skip it** (just do the task) for quick factual questions, single-step edits, and\nrequests already crisp enough to act on. A brief on a trivial ask is friction, not\nquality — proportionality is part of the skill. When in doubt, a one-line\nrestatement of the goal (\"I'll read this as X — say if not\") beats a full brief.\n\n## The brief\n\nCompile and show this exact shape — short, in plain business language:\n\n```\n**Task brief — [short task name]**\n- **Goal** — the deliverable, its format and its audience.\n- **Done when** — 2–5 checkable tests, one per line: \"When [condition],\n  [the output] [result]\", or just \"[the output] [result]\" where there is no\n  condition.\n- **Context** — everything known that bears on the task: the inputs\u002Ffiles\n  provided, relevant background from the conversation or memory, the entities \u002F\n  period \u002F parties involved, what already exists.\n- **Constraints** — what to avoid and the guardrails: task-specific limits\n  (scope, exclusions, deadline, tone) plus whichever standing rules materially\n  bite this task (see below).\n- **Assumptions** — anything filled in without asking, flagged so the user can\n  correct it with one word.\n- **Tooling** — the installed skill(s) or tool(s) that will do the work.\n```\n\n## Compiling it\n\n1. **Extract before you ask.** Mine the request, the conversation so far, any files\n   provided, and memory first — never ask the user for something already in front of\n   you.\n2. **Ask only what changes the output.** For each gap, test: would a different\n   answer produce different work? If yes, ask — **at most 2–3 questions**, via a\n   structured question tool where available, else as a short numbered list in chat.\n   If no, fill it with a sensible default and record it under **Assumptions**. Never\n   stack an interrogation in front of the task.\n3. **Make \"done\" checkable.** Turn the goal into the **Done when** tests:\n   - **One test per line** — split \"accurate and on time and on brand\" into three.\n   - **Swap vague words for something you could check** on delivery:\n\n     | Vague | Make it |\n     |---|---|\n     | \"something on\", \"a summary\", \"an overview\" | a named format and length — one-page memo, five talking points |\n     | \"quick\", \"high-level\", \"brief\" | a page or word cap |\n     | \"clean\", \"tidy\", \"fix\" | the specific defects removed |\n     | \"flag issues\", \"anything unusual\" | the threshold — variance above 5%, items over 30 days overdue |\n     | \"for the call\", \"for the board\" | the audience and the date |\n     | \"the latest numbers\" | the period and the source file |\n\n   - **Cover one failure case** where it matters — missing data, a threshold\n     breach, a figure that doesn't tie — and say what the output does then.\n   - **Five tests at most.** More means the task needs splitting.\n\n   A value you had to pick (the 5%, the page cap) is an assumption — list it under\n   **Assumptions** too.\n4. **Inject the standing constraints — don't ask about them.** Every organisation\n   has rules that apply to all its work; they go into **Constraints** automatically\n   (list only the ones that bite). Source them from the host environment first — the\n   system prompt, org instructions, a `CLAUDE.md` \u002F agents file, a principles or\n   style document. Where none are defined, apply these defaults: outputs are\n   **drafts for a person to review, not advice**; **never invent** a fact — ask or\n   leave a visibly flagged blank; keep spelling, date and currency conventions\n   consistent with the house style; **confidential data stays local** — de-identify\n   before anything crosses to an external tool; the **action boundary** — prepare,\n   never send \u002F post \u002F pay \u002F file \u002F sign without an explicit go-ahead.\n5. **Route the tooling.** Survey what is actually installed (skills, connectors,\n   scripts) and name the most specific one that will do the work. If a better-suited\n   capability exists but isn't installed, say so as advisory — never make the brief\n   depend on something the user doesn't have. If nothing fits, say that too, and\n   proceed with general capability. Multi-source research goes to `deep-research`\n   (if installed), which takes the confirmed brief as its plan input — no second\n   round of questions.\n6. **Pre-screen mode compatibility.** As part of the **Tooling** line, check the\n   routed skill against what the current environment actually supports — Python\n   helpers, credentials, network access, platform gates. If the skill needs\n   something the environment lacks, surface that in the brief (\"this route needs\n   X, which isn't available here — fallback is Y\"), so mode problems are visible\n   before work starts, not mid-task.\n7. **Confirm, then execute.** Show the brief and wait for a go \u002F corrections — a\n   one-word \"go\" is enough; this is a glance, not a form. Then do the work\n   **against the brief**.\n8. **Close the loop.** On delivery, run each **Done when** test and report it as\n   met or outstanding, plainly — never claim a check that didn't run.\n\n## Worked example (fictional)\n\n> User: *\"put together something on the Northwind account numbers for Thursday's\n> call\"*\n\nThe agent checks the folder (finds `northwind-q2-summary.xlsx`), then asks two\nquestions — audience (internal team vs the client) and form (one-page dashboard vs\ntalking-points memo) — and compiles:\n\n```\n**Task brief — Northwind Q2 numbers for Thursday's call**\n- **Goal** — a one-page dashboard (PDF via browser print) of the Northwind\n  account's Q2 numbers, for the internal team call on Thu 12 Mar.\n- **Done when**\n  - the page shows revenue vs budget, pipeline and overdue items, each with the\n    Q1 comparison;\n  - when revenue is more than 5% off budget, the dashboard flags it with a\n    one-line reason;\n  - when a figure is missing from the source file, it shows as a flagged blank,\n    not a zero;\n  - every figure traces to a cell in northwind-q2-summary.xlsx.\n- **Context** — source: northwind-q2-summary.xlsx (this folder); Q1 comparison\n  available in the same file; internal audience per your answer.\n- **Constraints** — internal draft, not for client distribution; figures traced\n  to the source file, none invented; consistent currency and date formats.\n- **Assumptions** — \"numbers\" = the account KPIs, not the contract terms (say\n  if you want those in); 5% as the budget-variance flag.\n- **Tooling** — the installed dashboard\u002Fvisualisation skill; spreadsheet read\n  via the local data engine.\n```\n\nUser replies \"go, add contract renewals\" — one correction, caught before the work,\nnot after.\n\n## Reusable-prompt mode\n\nUse it when the user asks for a prompt rather than a result (\"turn this into a\nreusable prompt\", \"improve my prompt so I can reuse it\", \"write a prompt for the\nteam\"), or when the task plainly recurs and the user agrees to save it. If the\nrecurring work is **research** (a quarterly market scan, a competitor screen) and\n`deep-research` is installed, save it as a deep-research recipe instead — it also holds\nthe preferred sources, fields and shelf lives. If the\nrecurring work needs scripts, templates or its own checks, it is a skill, not a\nprompt — say so and suggest building one.\n\n1. **Compile the brief as above** (steps 1–6), but write it for **any run**, not this\n   one: the things that change each run (period, account, source file) become named\n   inputs; the things that don't (format, thresholds, standing rules) are fixed in\n   the prompt.\n2. **Write the prompt** in this shape:\n\n   ```\n   # [Prompt name]\n   ## Role        — who the agent acts as, in one line\n   ## Inputs      — what the user supplies each run, as [square-bracket fields],\n   ##               and where each is usually found\n   ## Steps       — numbered; each ends with its own \"done when\"\n   ## Rules       — the standing rules that bite + task-specific limits\n   ## Output      — format, length, file name, where it is saved\n   ## Examples    — one normal run and one edge case (missing data, a breach),\n   ##               with concrete fictional values, not [fields]\n   ## Checks      — the Done-when tests, to run before handing over\n   ```\n\n3. **Save and show.** Save it as `[task-name]-prompt.md` in the working folder and\n   show it in chat. Don't run it unless the user asks. If they want it on a schedule,\n   say so and leave setting it up to their go-ahead.\n\n**Keep shareable prompts clean.** A saved prompt outlives the conversation and may be\npassed around: keep real client, counterparty and personal names and figures **out**\nof it — they belong in the inputs at run time, not in the prompt. Examples use\nfictional values.\n\n## Principles\n\n- **Drafts, not advice** — the brief and the work it frames are drafts for a person\n  to review, not professional advice.\n- **Never invent** — a gap the user doesn't close becomes a flagged assumption or a\n  visible blank, never a plausible guess.\n- **Proportionality** — the brief must cost less than the misunderstanding it\n  prevents; skip it on trivial asks.\n- **Honesty and calibration** — close the loop against the Goal line and report\n  what's met and what's outstanding, plainly.\n- **Checkable, not vague** — \"done\" is a short list of tests someone could tick or\n  cross on delivery, not an adjective.\n- **The confirmation moment is the value** — a misreading corrected before the work\n  starts is nearly free; the same correction after delivery costs the whole task.\n\n## Data handling\n\nThe brief itself often quotes client, counterparty or personal context — it lives\n**in chat and the local working folder only**; never send a brief to an external or\nthird-party tool. If a step in the *task* would cross to an external service,\nde-identify sensitive data first and confirm the egress with the user. This skill\nitself needs no network, no scripts and no credentials; reusable-prompt mode writes\none Markdown file to the local working folder.\n\n## Pitfalls\n\n1. **Interrogation fatigue** — more than 2–3 questions up front and users stop\n   answering; convert the rest to flagged assumptions.\n2. **Asking what you can read** — a question about something already in the folder\n   or the conversation erodes trust in the whole brief.\n3. **Generic constraints padding** — listing every standing rule on every brief\n   buries the one that matters; include only the rules that bite this task.\n4. **Skipping the close-the-loop step** — the Done-when tests are the test of\n   \"done\"; delivering without running them wastes the brief.\n5. **Tool dependence** — naming a tool the user doesn't have as the plan; keep\n   uninstalled capabilities advisory.\n6. **Adjectives as tests** — \"clear\", \"concise\", \"accurate\" can't be checked; swap\n   them for a format, a cap, a threshold or a source.\n7. **Real data in a saved prompt** — a reusable prompt gets shared; real names and\n   figures belong in the run-time inputs, not the file.\n\n## Feedback capture\n\nHave a correction or improvement to this skill? Capture it in a structured note —\nskill name, what you did, expected vs actual, severity, suggestion — save it as a\nplain text file, and hand it to the user to file. Manual, no fixed destination;\nfix in scope if asked. This keeps improvement signals reaching the skill author\nconsistently instead of evaporating in chat.\n\n## Verification checklist\n\n- [ ] Substantial task confirmed (or explicit invocation) — not a trivial ask.\n- [ ] Request, conversation, files and memory mined before any question was asked.\n- [ ] At most 2–3 questions asked, each one output-changing.\n- [ ] 2–5 Done-when tests, one test per line, no vague words; picked thresholds\n      listed as assumptions.\n- [ ] Standing rules injected from the host environment (or the defaults), not asked.\n- [ ] Routed skill pre-screened against what the environment supports.\n- [ ] Brief shown and confirmed by the user before the work started.\n- [ ] Each Done-when test run on delivery; met vs outstanding reported.\n- [ ] Reusable-prompt mode: run-specific values are inputs, examples fictional,\n      file saved locally and not run unless asked.\n\n## Related work\n\nThe Done-when tests borrow the one-condition-one-result shape of EARS (the Easy\nApproach to Requirements Syntax, Mavin et al., 2009), and the reusable-prompt mode the\nRole \u002F Steps \u002F Examples \u002F Output layout — both ideas met in the `prompt-optimizer`\nskill in daymade\u002Fclaude-code-skills. Ideas only; no text or code reused.\n\n## Requirements\n\n- None — a reasoning-only skill: no Python, no network, no credentials. Runs in any\n  agent environment that can read this file.\n","# Task Brief\n\nThin requests are the biggest cause of mediocre agent output: the agent guesses the\ndeliverable, the audience and the guardrails, and the user only discovers the\nmisreading after the work is done. This skill moves that discovery to the **cheapest\npossible point** — before the work starts — by having the agent compile a short\n**task brief**, show it, and only then execute. The user never has to write a better\nprompt; the agent does the structuring.\n\nIt has two modes:\n- **Brief and run** (default) — compile the brief, confirm, then do the work.\n- **Reusable prompt** — compile the same brief, but save it as a structured prompt\n  for a task that recurs (a monthly report, a scheduled job, a prompt handed to a\n  colleague) instead of running it. See [Reusable-prompt mode](#reusable-prompt-mode).\n\n## When to run it — and when not to\n\n**Run it when:**\n- the user explicitly asks (\"brief this first\", \"make sure you understand before starting\"); or\n- the task is **substantial** — multi-step, multi-document, or a deliverable someone\n  else will read (a memo, a report, a dashboard, a document batch) — and the request\n  is thin or ambiguous enough that two reasonable readings would produce different work.\n\n**Skip it** (just do the task) for quick factual questions, single-step edits, and\nrequests already crisp enough to act on. A brief on a trivial ask is friction, not\nquality — proportionality is part of the skill. When in doubt, a one-line\nrestatement of the goal (\"I'll read this as X — say if not\") beats a full brief.\n\n## The brief\n\nCompile and show this exact shape — short, in plain business language:\n\n```\n**Task brief — [short task name]**\n- **Goal** — the deliverable, its format and its audience.\n- **Done when** — 2–5 checkable tests, one per line: \"When [condition],\n  [the output] [result]\", or just \"[the output] [result]\" where there is no\n  condition.\n- **Context** — everything known that bears on the task: the inputs\u002Ffiles\n  provided, relevant background from the conversation or memory, the entities \u002F\n  period \u002F parties involved, what already exists.\n- **Constraints** — what to avoid and the guardrails: task-specific limits\n  (scope, exclusions, deadline, tone) plus whichever standing rules materially\n  bite this task (see below).\n- **Assumptions** — anything filled in without asking, flagged so the user can\n  correct it with one word.\n- **Tooling** — the installed skill(s) or tool(s) that will do the work.\n```\n\n## Compiling it\n\n1. **Extract before you ask.** Mine the request, the conversation so far, any files\n   provided, and memory first — never ask the user for something already in front of\n   you.\n2. **Ask only what changes the output.** For each gap, test: would a different\n   answer produce different work? If yes, ask — **at most 2–3 questions**, via a\n   structured question tool where available, else as a short numbered list in chat.\n   If no, fill it with a sensible default and record it under **Assumptions**. Never\n   stack an interrogation in front of the task.\n3. **Make \"done\" checkable.** Turn the goal into the **Done when** tests:\n   - **One test per line** — split \"accurate and on time and on brand\" into three.\n   - **Swap vague words for something you could check** on delivery:\n\n     | Vague | Make it |\n     |---|---|\n     | \"something on\", \"a summary\", \"an overview\" | a named format and length — one-page memo, five talking points |\n     | \"quick\", \"high-level\", \"brief\" | a page or word cap |\n     | \"clean\", \"tidy\", \"fix\" | the specific defects removed |\n     | \"flag issues\", \"anything unusual\" | the threshold — variance above 5%, items over 30 days overdue |\n     | \"for the call\", \"for the board\" | the audience and the date |\n     | \"the latest numbers\" | the period and the source file |\n\n   - **Cover one failure case** where it matters — missing data, a threshold\n     breach, a figure that doesn't tie — and say what the output does then.\n   - **Five tests at most.** More means the task needs splitting.\n\n   A value you had to pick (the 5%, the page cap) is an assumption — list it under\n   **Assumptions** too.\n4. **Inject the standing constraints — don't ask about them.** Every organisation\n   has rules that apply to all its work; they go into **Constraints** automatically\n   (list only the ones that bite). Source them from the host environment first — the\n   system prompt, org instructions, a `CLAUDE.md` \u002F agents file, a principles or\n   style document. Where none are defined, apply these defaults: outputs are\n   **drafts for a person to review, not advice**; **never invent** a fact — ask or\n   leave a visibly flagged blank; keep spelling, date and currency conventions\n   consistent with the house style; **confidential data stays local** — de-identify\n   before anything crosses to an external tool; the **action boundary** — prepare,\n   never send \u002F post \u002F pay \u002F file \u002F sign without an explicit go-ahead.\n5. **Route the tooling.** Survey what is actually installed (skills, connectors,\n   scripts) and name the most specific one that will do the work. If a better-suited\n   capability exists but isn't installed, say so as advisory — never make the brief\n   depend on something the user doesn't have. If nothing fits, say that too, and\n   proceed with general capability. Multi-source research goes to `deep-research`\n   (if installed), which takes the confirmed brief as its plan input — no second\n   round of questions.\n6. **Pre-screen mode compatibility.** As part of the **Tooling** line, check the\n   routed skill against what the current environment actually supports — Python\n   helpers, credentials, network access, platform gates. If the skill needs\n   something the environment lacks, surface that in the brief (\"this route needs\n   X, which isn't available here — fallback is Y\"), so mode problems are visible\n   before work starts, not mid-task.\n7. **Confirm, then execute.** Show the brief and wait for a go \u002F corrections — a\n   one-word \"go\" is enough; this is a glance, not a form. Then do the work\n   **against the brief**.\n8. **Close the loop.** On delivery, run each **Done when** test and report it as\n   met or outstanding, plainly — never claim a check that didn't run.\n\n## Worked example (fictional)\n\n> User: *\"put together something on the Northwind account numbers for Thursday's\n> call\"*\n\nThe agent checks the folder (finds `northwind-q2-summary.xlsx`), then asks two\nquestions — audience (internal team vs the client) and form (one-page dashboard vs\ntalking-points memo) — and compiles:\n\n```\n**Task brief — Northwind Q2 numbers for Thursday's call**\n- **Goal** — a one-page dashboard (PDF via browser print) of the Northwind\n  account's Q2 numbers, for the internal team call on Thu 12 Mar.\n- **Done when**\n  - the page shows revenue vs budget, pipeline and overdue items, each with the\n    Q1 comparison;\n  - when revenue is more than 5% off budget, the dashboard flags it with a\n    one-line reason;\n  - when a figure is missing from the source file, it shows as a flagged blank,\n    not a zero;\n  - every figure traces to a cell in northwind-q2-summary.xlsx.\n- **Context** — source: northwind-q2-summary.xlsx (this folder); Q1 comparison\n  available in the same file; internal audience per your answer.\n- **Constraints** — internal draft, not for client distribution; figures traced\n  to the source file, none invented; consistent currency and date formats.\n- **Assumptions** — \"numbers\" = the account KPIs, not the contract terms (say\n  if you want those in); 5% as the budget-variance flag.\n- **Tooling** — the installed dashboard\u002Fvisualisation skill; spreadsheet read\n  via the local data engine.\n```\n\nUser replies \"go, add contract renewals\" — one correction, caught before the work,\nnot after.\n\n## Reusable-prompt mode\n\nUse it when the user asks for a prompt rather than a result (\"turn this into a\nreusable prompt\", \"improve my prompt so I can reuse it\", \"write a prompt for the\nteam\"), or when the task plainly recurs and the user agrees to save it. If the\nrecurring work is **research** (a quarterly market scan, a competitor screen) and\n`deep-research` is installed, save it as a deep-research recipe instead — it also holds\nthe preferred sources, fields and shelf lives. If the\nrecurring work needs scripts, templates or its own checks, it is a skill, not a\nprompt — say so and suggest building one.\n\n1. **Compile the brief as above** (steps 1–6), but write it for **any run**, not this\n   one: the things that change each run (period, account, source file) become named\n   inputs; the things that don't (format, thresholds, standing rules) are fixed in\n   the prompt.\n2. **Write the prompt** in this shape:\n\n   ```\n   # [Prompt name]\n   ## Role        — who the agent acts as, in one line\n   ## Inputs      — what the user supplies each run, as [square-bracket fields],\n   ##               and where each is usually found\n   ## Steps       — numbered; each ends with its own \"done when\"\n   ## Rules       — the standing rules that bite + task-specific limits\n   ## Output      — format, length, file name, where it is saved\n   ## Examples    — one normal run and one edge case (missing data, a breach),\n   ##               with concrete fictional values, not [fields]\n   ## Checks      — the Done-when tests, to run before handing over\n   ```\n\n3. **Save and show.** Save it as `[task-name]-prompt.md` in the working folder and\n   show it in chat. Don't run it unless the user asks. If they want it on a schedule,\n   say so and leave setting it up to their go-ahead.\n\n**Keep shareable prompts clean.** A saved prompt outlives the conversation and may be\npassed around: keep real client, counterparty and personal names and figures **out**\nof it — they belong in the inputs at run time, not in the prompt. Examples use\nfictional values.\n\n## Principles\n\n- **Drafts, not advice** — the brief and the work it frames are drafts for a person\n  to review, not professional advice.\n- **Never invent** — a gap the user doesn't close becomes a flagged assumption or a\n  visible blank, never a plausible guess.\n- **Proportionality** — the brief must cost less than the misunderstanding it\n  prevents; skip it on trivial asks.\n- **Honesty and calibration** — close the loop against the Goal line and report\n  what's met and what's outstanding, plainly.\n- **Checkable, not vague** — \"done\" is a short list of tests someone could tick or\n  cross on delivery, not an adjective.\n- **The confirmation moment is the value** — a misreading corrected before the work\n  starts is nearly free; the same correction after delivery costs the whole task.\n\n## Data handling\n\nThe brief itself often quotes client, counterparty or personal context — it lives\n**in chat and the local working folder only**; never send a brief to an external or\nthird-party tool. If a step in the *task* would cross to an external service,\nde-identify sensitive data first and confirm the egress with the user. This skill\nitself needs no network, no scripts and no credentials; reusable-prompt mode writes\none Markdown file to the local working folder.\n\n## Pitfalls\n\n1. **Interrogation fatigue** — more than 2–3 questions up front and users stop\n   answering; convert the rest to flagged assumptions.\n2. **Asking what you can read** — a question about something already in the folder\n   or the conversation erodes trust in the whole brief.\n3. **Generic constraints padding** — listing every standing rule on every brief\n   buries the one that matters; include only the rules that bite this task.\n4. **Skipping the close-the-loop step** — the Done-when tests are the test of\n   \"done\"; delivering without running them wastes the brief.\n5. **Tool dependence** — naming a tool the user doesn't have as the plan; keep\n   uninstalled capabilities advisory.\n6. **Adjectives as tests** — \"clear\", \"concise\", \"accurate\" can't be checked; swap\n   them for a format, a cap, a threshold or a source.\n7. **Real data in a saved prompt** — a reusable prompt gets shared; real names and\n   figures belong in the run-time inputs, not the file.\n\n## Feedback capture\n\nHave a correction or improvement to this skill? Capture it in a structured note —\nskill name, what you did, expected vs actual, severity, suggestion — save it as a\nplain text file, and hand it to the user to file. Manual, no fixed destination;\nfix in scope if asked. This keeps improvement signals reaching the skill author\nconsistently instead of evaporating in chat.\n\n## Verification checklist\n\n- [ ] Substantial task confirmed (or explicit invocation) — not a trivial ask.\n- [ ] Request, conversation, files and memory mined before any question was asked.\n- [ ] At most 2–3 questions asked, each one output-changing.\n- [ ] 2–5 Done-when tests, one test per line, no vague words; picked thresholds\n      listed as assumptions.\n- [ ] Standing rules injected from the host environment (or the defaults), not asked.\n- [ ] Routed skill pre-screened against what the environment supports.\n- [ ] Brief shown and confirmed by the user before the work started.\n- [ ] Each Done-when test run on delivery; met vs outstanding reported.\n- [ ] Reusable-prompt mode: run-specific values are inputs, examples fictional,\n      file saved locally and not run unless asked.\n\n## Related work\n\nThe Done-when tests borrow the one-condition-one-result shape of EARS (the Easy\nApproach to Requirements Syntax, Mavin et al., 2009), and the reusable-prompt mode the\nRole \u002F Steps \u002F Examples \u002F Output layout — both ideas met in the `prompt-optimizer`\nskill in daymade\u002Fclaude-code-skills. Ideas only; no text or code reused.\n\n## Requirements\n\n- None — a reasoning-only skill: no Python, no network, no credentials. Runs in any\n  agent environment that can read this file.\n",{"name":5,"license":14,"metadata":23,"description":7},{"tags":24,"author":9,"version":30,"platforms":31},[25,26,27,28,29],"planning","brief","scoping","productivity","prompts","1.3.0",[32,33,34],"linux","macos","windows","main","e943e3285cd03c44b2c762a9a52328e3d16444b1","productivity\u002Ftask-brief",22,"https:\u002F\u002Fgithub.com\u002Fmoonlight-lupin\u002Fagent-skills","https:\u002F\u002Fhermes-agent.nousresearch.com","https:\u002F\u002Fgithub.com\u002Fmoonlight-lupin","Accounting\u002FFP&A\u002FM&A\u002FPERE guy by day, AI\u002Fself-host\u002Fprogramming hobbyist by night. \r\n\r\nPassionate in solving practical business issues.","2026-11-13 19:46:18",{"source":8,"skillName":5,"commands":45,"prompt":55,"zipAvailable":56},[46,49,52],{"tool":47,"command":48},"npx","npx skills add moonlight-lupin\u002Fagent-skills --skill task-brief",{"tool":50,"command":51},"bunx","bunx skills add moonlight-lupin\u002Fagent-skills --skill task-brief",{"tool":53,"command":54},"pnpm","pnpm dlx skills add moonlight-lupin\u002Fagent-skills --skill task-brief","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fmoonlight-lupin\u002Fagent-skills\u002Fblob\u002Fmain\u002Fproductivity\u002Ftask-brief\u002FSKILL.md",true,[58],{"path":59,"resourceType":60,"resourceTypeDesc":61,"sizeBytes":62,"sourceHtmlUrl":63},"README.md","REFERENCE","参考文档",1337,"https:\u002F\u002Fgithub.com\u002Fmoonlight-lupin\u002Fagent-skills\u002Fblob\u002Fmain\u002Fproductivity\u002Ftask-brief\u002FREADME.md",[65,72,79,86,93,100],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":18},"01M78ATNRV7RYQ8M314V0QZX18","website-scraping","Generic playbook for extracting structured data from any website — hotel prices, flight fares, e-commerce listings, real estate, jobs, competitor product catalogues, anything where the goal is to turn one or more URLs into clean records on disk. Use this skill whenever the user mentions scraping, harvesting, extracting, or pulling data from a website; names a specific site or competitor they want data from; asks how to handle JavaScript-rendered pages, Cloudflare blocks, bot detection, Turnstile challenges, or Playwright; wants to monitor prices over time; needs to automate a copy-paste research task; or says things like \"just get the data from X\" or \"I need a script that grabs N from site Y\". Covers recon, picking the lightest extraction tool that works, surviving anti-bot defences, and writing clean JSONL output with a run manifest. Scope is scraping only — landing data in a database, scheduling, and downstream pipeline work are explicitly out of scope and live in other tools.","https:\u002F\u002Fgithub.com\u002Fmoonlight-lupin\u002Fagent-skills\u002Fblob\u002Fmain\u002Fweb-scraping\u002Fwebsite-scraping\u002FSKILL.md",[],[],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":18},"01M78ATK9XGEKSH3QRN7P2MEER","greet","Greets a named person warmly.","https:\u002F\u002Fgithub.com\u002Fmoonlight-lupin\u002Fagent-skills\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fclaude_plugin_sample\u002Fskills\u002Fgreet\u002FSKILL.md",[],[],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":18},"01M78ATEAQB5CK2D4MKPCW4YHQ","skill-quality-review","Use when auditing or reviewing skill quality — batch checks across a library, deep per-skill grep methodology, usage stats, efficacy testing, or fix planning. Single consolidated home for the writing-for-agents review framework and the 3-surface review methodology.","https:\u002F\u002Fgithub.com\u002Fmoonlight-lupin\u002Fagent-skills\u002Fblob\u002Fmain\u002Fsoftware-development\u002Fskill-quality-review\u002FSKILL.md",[],[],{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":18},"01M78AT58EM57023QSC79S8R6D","people-enrichment","Use when the user has a list of names or companies and wants their job, employer, work history, LinkedIn URLs or company profiles looked up, or wants to find people or companies matching criteria — trigger on phrases like \"enrich these contacts\", \"find their LinkedIn\", \"look up these people\", \"who works at X\", \"find directors at Y\", \"enrich these companies\", or \"get firmographics for\". Enrich and search People Data Labs (PDL) Person and Company data, writing results to .xlsx. Five operations: enrich named PEOPLE into a profile, employment history and LinkedIn URL; IDENTIFY an ambiguous person as several scored candidates; SEARCH people by criteria (company, title, location); enrich COMPANIES into firmographics (industry, size, employees, HQ, founded, LinkedIn); and SEARCH companies by criteria. Do not scrape LinkedIn directly — this uses a licensed data aggregator (PDL) instead, for legal and reliability reasons.","https:\u002F\u002Fgithub.com\u002Fmoonlight-lupin\u002Fagent-skills\u002Fblob\u002Fmain\u002Fresearch\u002Fpeople-enrichment\u002FSKILL.md",[],[],{"skillNo":94,"name":95,"nameZh":6,"description":96,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":97,"categories":98,"keywords":99,"publishedAt":18},"01M78AT0AZJ7WFY73766NBA1EC","notebooklm-mode","Source-grounded research pipeline inspired by NotebookLM. Collect verbatim source extracts into a temporary workspace, then answer questions and create deliverables grounded in those sources. Two grounding modes: strict (vault-only) and augmented (vault + labeled [background] knowledge). Trigger when the user asks to research a topic from sources, wants NotebookLM-style grounded answers, asks to build a resource vault, or says \"notebooklm mode\".","https:\u002F\u002Fgithub.com\u002Fmoonlight-lupin\u002Fagent-skills\u002Fblob\u002Fmain\u002Fresearch\u002Fnotebooklm-mode\u002FSKILL.md",[],[],{"skillNo":101,"name":102,"nameZh":6,"description":103,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":104,"categories":105,"keywords":106,"publishedAt":18},"01M78ASTJVZ9G5JY6GM40SF909","news-monitoring","Use when a topic or news stream needs recurring monitoring — cover it with web search, multi-language sources, digest formatting, and automated delivery via Hermes cron jobs. Covers search strategy, source selection, Chinese-language platforms, and digest templates.","https:\u002F\u002Fgithub.com\u002Fmoonlight-lupin\u002Fagent-skills\u002Fblob\u002Fmain\u002Fresearch\u002Fnews-monitoring\u002FSKILL.md",[],[]]