{
  "version": 1,
  "exportedAt": "2026-10-13T03:58:23.620Z",
  "prompts": [
    {
      "id": "56b9949f-bc5e-4abc-b250-d9e8cbcf3f55",
      "title": "Summarize for an audience",
      "body": "Summarize the following {{topic}} for a {{audience}} in {{length}} words.\n\nFocus on what matters to that audience. Use plain language and short paragraphs.\n\nTEXT:\n\"\"\"\n{{input}}\n\"\"\"",
      "description": "Condenses any long text for a specific audience at a target length.",
      "category": "writing",
      "targetModels": [
        "GPT-4o",
        "Claude"
      ],
      "tags": [
        "general",
        "summarization"
      ],
      "isFavorite": true,
      "usageCount": 12,
      "lastUsedAt": "2026-08-26T12:38:25.028Z",
      "createdAt": "2026-08-26T14:38:25.028Z",
      "updatedAt": "2026-08-26T14:38:25.028Z"
    },
    {
      "id": "ce57f188-910a-4cf9-a3e2-34492c1dcc3f",
      "title": "Ruthless code review",
      "body": "You are a staff engineer reviewing a pull request in {{language}}.\n\nReview the code below for:\n1. Correctness bugs (rank P0/P1/P2)\n2. Security issues\n3. Readability and naming\n4. Missing tests\n\nBe direct. No praise. End with the single most important change to make.\n\n```{{language}}\n{{code}}\n```",
      "description": "Senior-engineer code review: correctness, security, style — ranked by severity.",
      "category": "coding",
      "targetModels": [
        "Claude",
        "GPT-4.1"
      ],
      "tags": [
        "code-review",
        "engineering"
      ],
      "isFavorite": true,
      "usageCount": 21,
      "lastUsedAt": "2026-08-26T12:38:25.524Z",
      "createdAt": "2026-08-26T14:38:25.524Z",
      "updatedAt": "2026-08-26T14:38:25.524Z"
    },
    {
      "id": "f90cf07d-1e2d-4cb2-91d7-8c1083c73b7f",
      "title": "Blog outline generator",
      "body": "Create a blog post outline for the topic \"{{topic}}\".\n\nAudience: {{audience}}\nTone: {{tone}}\nTarget length: {{length}} words\n\nReturn:\n- 5 candidate titles (one curiosity-gap, one how-to, one listicle)\n- H2/H3 outline with 1-line notes per section\n- Suggested internal-link anchor points",
      "description": "SEO-aware outline with H2/H3 structure and hook angles.",
      "category": "marketing",
      "targetModels": [
        "GPT-4o",
        "Gemini"
      ],
      "tags": [
        "blog",
        "content",
        "seo"
      ],
      "isFavorite": false,
      "usageCount": 7,
      "lastUsedAt": "2026-08-26T12:38:26.006Z",
      "createdAt": "2026-08-26T14:38:26.006Z",
      "updatedAt": "2026-08-26T14:38:26.006Z"
    },
    {
      "id": "091f684f-a493-4391-8818-0331935bae7a",
      "title": "Structured data extraction",
      "body": "Extract the following fields from the text below and return ONLY valid JSON (no markdown fence):\n\n{\n  \"person\": string | null,\n  \"company\": string | null,\n  \"date\": \"YYYY-MM-DD\" | null,\n  \"amount_usd\": number | null,\n  \"sentiment\": \"positive\" | \"neutral\" | \"negative\"\n}\n\nIf a field is not present, use null. Do not guess.\n\nTEXT:\n{{text}}",
      "description": "Extracts typed fields from messy text into strict JSON.",
      "category": "research",
      "targetModels": [
        "GPT-4o",
        "Claude"
      ],
      "tags": [
        "extraction",
        "json"
      ],
      "isFavorite": true,
      "usageCount": 15,
      "lastUsedAt": "2026-08-26T12:38:26.680Z",
      "createdAt": "2026-08-26T14:38:26.680Z",
      "updatedAt": "2026-08-26T14:38:26.680Z"
    },
    {
      "id": "02fb9fb4-8cd5-4485-8817-3bdd4860d77c",
      "title": "System prompt: terse Linux expert",
      "body": "You are a Linux systems expert. Answer with the exact command or config first, then at most 2 lines of explanation. Never use GUI instructions. Assume Debian-based distros unless told otherwise. If the request is ambiguous, ask one clarifying question — otherwise proceed.",
      "description": "Base system prompt for a no-fluff terminal assistant.",
      "category": "system",
      "targetModels": [
        "Claude",
        "Llama",
        "Mistral"
      ],
      "tags": [
        "persona",
        "system-prompt"
      ],
      "isFavorite": false,
      "usageCount": 4,
      "lastUsedAt": "2026-08-26T12:38:27.162Z",
      "createdAt": "2026-08-26T14:38:27.162Z",
      "updatedAt": "2026-08-26T14:38:27.162Z"
    },
    {
      "id": "74902240-4552-47ca-b195-08e0dd658f58",
      "title": "Ad copy variants",
      "body": "Write {{count}} ad copy variants for {{product}}.\n\nTarget: {{audience}}\nPlatform: {{platform}}\nBrand voice: {{tone}}\nConstraint: max {{char_limit}} characters per variant.\n\nUse four different hooks across variants: pain-point, aspiration, social proof, and FOMO. Label each variant with its hook.",
      "description": "Generates ad variants across hooks (pain, aspiration, proof, FOMO).",
      "category": "marketing",
      "targetModels": [
        "GPT-4o",
        "Gemini"
      ],
      "tags": [
        "ads",
        "copywriting"
      ],
      "isFavorite": false,
      "usageCount": 3,
      "lastUsedAt": "2026-08-26T12:38:27.644Z",
      "createdAt": "2026-08-26T14:38:27.644Z",
      "updatedAt": "2026-08-26T14:38:27.644Z"
    },
    {
      "id": "bb831f50-ef67-4e30-8627-c4533e1c2c85",
      "title": "Unit test scaffolder",
      "body": "Write {{framework}} unit tests for the function below.\n\nRequirements:\n- Table-driven style (cases array)\n- Cover: happy path, boundary values, empty input, unicode, and one adversarial case\n- Do not modify the function itself\n\n```\n{{code}}\n```",
      "description": "Generates table-driven unit tests for a function.",
      "category": "coding",
      "targetModels": [
        "GPT-4.1",
        "Claude"
      ],
      "tags": [
        "code-generation",
        "testing"
      ],
      "isFavorite": false,
      "usageCount": 9,
      "lastUsedAt": "2026-08-26T12:38:28.126Z",
      "createdAt": "2026-08-26T14:38:28.126Z",
      "updatedAt": "2026-08-26T14:38:28.126Z"
    },
    {
      "id": "85454185-aee9-4a0e-8561-c0a7a2c04041",
      "title": "Image prompt: product hero shot",
      "body": "A professional product photograph of {{product}}, on a {{surface}} surface with a {{background}} background, {{lighting}} lighting, shot at {{angle}}, 85mm lens, f/2.8, high detail, commercial quality, no text or watermarks.",
      "description": "Structured image-gen prompt with camera and lighting language.",
      "category": "image",
      "targetModels": [
        "Gemini",
        "Other"
      ],
      "tags": [
        "ecommerce",
        "image"
      ],
      "isFavorite": false,
      "usageCount": 2,
      "lastUsedAt": "2026-08-26T12:38:28.607Z",
      "createdAt": "2026-08-26T14:38:28.607Z",
      "updatedAt": "2026-08-26T14:38:28.607Z"
    },
    {
      "id": "bf0e9d8b-e7de-4f45-b59e-5bd208b56374",
      "title": "Competitive teardown",
      "body": "Analyze the competitor \"{{competitor}}\" in the {{industry}} market.\n\nBased on the material below, produce:\n1. Positioning statement (their one-liner)\n2. Pricing strategy inference\n3. Their 3 strongest moats\n4. Their 3 weakest points\n5. One wedge we could exploit\n\nMATERIAL:\n{{material}}",
      "description": "Structured competitor analysis from their marketing site.",
      "category": "analysis",
      "targetModels": [
        "Claude",
        "GPT-4o"
      ],
      "tags": [
        "research",
        "strategy"
      ],
      "isFavorite": false,
      "usageCount": 6,
      "lastUsedAt": "2026-08-26T12:38:29.090Z",
      "createdAt": "2026-08-26T14:38:29.090Z",
      "updatedAt": "2026-08-26T14:38:29.090Z"
    },
    {
      "id": "f7d134c0-87a0-428c-8405-b9c3d1e9b1b4",
      "title": "Rewrite in plain English",
      "body": "Rewrite the text below in plain English at a grade-{{grade_level}} reading level.\n\nRules:\n- Keep every fact\n- Sentences ≤ 20 words\n- No jargon unless you define it inline\n- Keep the original tone where possible\n\nTEXT:\n{{text}}",
      "description": "Simplifies jargon-heavy text to grade-8 reading level.",
      "category": "writing",
      "targetModels": [
        "GPT-4o",
        "Gemini",
        "Claude"
      ],
      "tags": [
        "editing",
        "plain-language"
      ],
      "isFavorite": true,
      "usageCount": 11,
      "lastUsedAt": "2026-08-26T12:38:29.571Z",
      "createdAt": "2026-08-26T14:38:29.571Z",
      "updatedAt": "2026-08-26T14:38:29.571Z"
    },
    {
      "id": "b5257a7b-d1e3-4268-b4e6-3fe3823be20b",
      "title": "SQL query from a question",
      "body": "Given this schema:\n\n{{schema}}\n\nWrite a single PostgreSQL query answering: \"{{question}}\"\n\nRules: SELECT statements only, use CTEs for readability, comment each CTE in one line.",
      "description": "Translates a natural-language question into a safe SELECT.",
      "category": "coding",
      "targetModels": [
        "GPT-4.1"
      ],
      "tags": [
        "data",
        "sql"
      ],
      "isFavorite": false,
      "usageCount": 5,
      "lastUsedAt": "2026-08-26T12:38:30.055Z",
      "createdAt": "2026-08-26T14:38:30.055Z",
      "updatedAt": "2026-08-26T14:38:30.055Z"
    },
    {
      "id": "e673b0c4-f61e-469a-9119-e9b8460081d9",
      "title": "Meeting notes → action items",
      "body": "Convert the meeting transcript below into:\n\n1. DECISIONS (bullet list, one line each)\n2. ACTION ITEMS (table: task | owner | due)\n3. OPEN QUESTIONS (unresolved threads)\n\nIf an owner is unclear, write \"UNASSIGNED\". Do not invent tasks.\n\nTRANSCRIPT:\n{{transcript}}",
      "description": "Converts raw transcript into decisions + owned action items.",
      "category": "analysis",
      "targetModels": [
        "Claude",
        "GPT-4o"
      ],
      "tags": [
        "meetings",
        "productivity"
      ],
      "isFavorite": false,
      "usageCount": 8,
      "lastUsedAt": "2026-08-26T12:38:30.540Z",
      "createdAt": "2026-08-26T14:38:30.540Z",
      "updatedAt": "2026-08-26T14:38:30.540Z"
    },
    {
      "id": "945e7a0e-0e99-4833-a1fd-2289b601eb9f",
      "title": "Cold email that doesn't suck",
      "body": "Write a cold email to a {{role}} at a {{company_type}} company.\n\nOffer: {{offer}}\nTrigger observation: {{trigger}}\n\nRules: exactly 3 sentences, no \"I hope this finds you well\", one clear CTA, subject line ≤ 6 words.",
      "description": "3-sentence cold outreach with a specific trigger.",
      "category": "marketing",
      "targetModels": [
        "GPT-4o"
      ],
      "tags": [
        "email",
        "sales"
      ],
      "isFavorite": false,
      "usageCount": 2,
      "lastUsedAt": "2026-08-26T12:38:31.023Z",
      "createdAt": "2026-08-26T14:38:31.023Z",
      "updatedAt": "2026-08-26T14:38:31.023Z"
    },
    {
      "id": "7038334d-9ce1-41da-94ca-b2796168bf76",
      "title": "Few-shot classifier",
      "body": "You are a {{domain}} classifier. Classify each input into exactly one of: {{labels}}.\n\nExamples:\n{{examples}}\n\nNow classify the following. Reply with ONLY the label, nothing else.\n\nINPUT: {{input}}",
      "description": "Template for building few-shot classification prompts.",
      "category": "system",
      "targetModels": [
        "GPT-4o",
        "Claude",
        "Llama"
      ],
      "tags": [
        "classification",
        "few-shot"
      ],
      "isFavorite": true,
      "usageCount": 14,
      "lastUsedAt": "2026-08-26T12:38:31.504Z",
      "createdAt": "2026-08-26T14:38:31.504Z",
      "updatedAt": "2026-08-26T14:38:31.504Z"
    },
    {
      "id": "6ee36076-a8c2-47ae-90ef-a28ac2386b01",
      "title": "API design reviewer",
      "body": "Review this REST API contract for developer experience:\n\n{{contract}}\n\nCheck: naming consistency, status-code discipline, pagination, error envelope, versioning, and idempotency of mutations. Output a table of issues with severity and a suggested fix.",
      "description": "Critiques a REST API contract for consistency and DX.",
      "category": "coding",
      "targetModels": [
        "Claude"
      ],
      "tags": [
        "api",
        "review"
      ],
      "isFavorite": false,
      "usageCount": 1,
      "lastUsedAt": "2026-08-26T12:38:31.985Z",
      "createdAt": "2026-08-26T14:38:31.985Z",
      "updatedAt": "2026-08-26T14:38:31.985Z"
    },
    {
      "id": "564d7ca0-29a2-4c45-aa1a-14c5cdf65e52",
      "title": "Reading list curator",
      "body": "Build a {{duration}} learning path for {{topic}}.\n\nMy background: {{background}}\nGoal: {{goal}}\n\nFormat: ordered list. Each item: resource type, why this order, and estimated hours. End with one project that proves the skill.",
      "description": "Builds a prioritized learning path with time estimates.",
      "category": "research",
      "targetModels": [
        "Gemini",
        "GPT-4o"
      ],
      "tags": [
        "curation",
        "learning"
      ],
      "isFavorite": false,
      "usageCount": 3,
      "lastUsedAt": "2026-08-26T12:38:32.465Z",
      "createdAt": "2026-08-26T14:38:32.465Z",
      "updatedAt": "2026-08-26T14:38:32.465Z"
    },
    {
      "id": "cd642566-abda-4c75-a535-0dd38e93a1c9",
      "title": "Release notes from commits",
      "body": "Turn the commit log below into release notes for {{product}} v{{version}}.\n\nAudience: {{audience}}\n\nGroup into: ✨ New · 🐛 Fixed · ⚡ Improved. Skip internal refactors. One line per item, present tense.",
      "description": "Turns a git log into user-facing release notes.",
      "category": "writing",
      "targetModels": [
        "GPT-4o",
        "Claude"
      ],
      "tags": [
        "changelog",
        "developer-relations"
      ],
      "isFavorite": false,
      "usageCount": 6,
      "lastUsedAt": "2026-08-26T12:38:32.946Z",
      "createdAt": "2026-08-26T14:38:32.946Z",
      "updatedAt": "2026-08-26T14:38:32.946Z"
    },
    {
      "id": "30a722b7-d3ec-44b6-b65e-6b1d124b7f91",
      "title": "Character consistency sheet",
      "body": "Maintain this character sheet while writing {{scene}}.\n\nCHARACTER:\n{{character}}\n\nRules:\n- Never break the voice defined above\n- Keep responses ≤ {{length}} words\n- End every scene beat on the character's dominant emotion",
      "description": "Locks a persona's voice for long-form fiction.",
      "category": "writing",
      "targetModels": [
        "Claude",
        "GPT-4o"
      ],
      "tags": [
        "fiction",
        "persona"
      ],
      "isFavorite": false,
      "usageCount": 2,
      "lastUsedAt": "2026-08-26T12:38:33.428Z",
      "createdAt": "2026-08-26T14:38:33.428Z",
      "updatedAt": "2026-08-26T14:38:33.428Z"
    },
    {
      "id": "83c5b306-738f-4525-b2dc-1eb63deced60",
      "title": "Incident postmortem skeleton",
      "body": "Draft a blameless postmortem for incident \"{{incident}}\".\n\nRaw timeline:\n{{timeline}}\n\nSections: Summary · Impact (users/duration) · Timeline · Root cause · What went well · What went wrong · Action items (owner + due). Be factual; no blame.",
      "description": "Blameless postmortem structure filled from raw timeline.",
      "category": "analysis",
      "targetModels": [
        "GPT-4o"
      ],
      "tags": [
        "incident",
        "sre"
      ],
      "isFavorite": false,
      "usageCount": 1,
      "lastUsedAt": "2026-08-26T12:38:33.814Z",
      "createdAt": "2026-08-26T14:38:33.814Z",
      "updatedAt": "2026-08-26T14:38:33.814Z"
    },
    {
      "id": "8238b633-cf11-4e53-88d6-30ac3217b527",
      "title": "Landing page hero copy",
      "body": "Write 5 hero-section pairs (headline ≤ 10 words, subhead ≤ 25 words) for {{product}}.\n\nWho it's for: {{audience}}\nPrimary pain: {{pain}}\nDifferentiator: {{wedge}}\n\nRank them by clarity first, cleverness second. Mark the winner and explain the ranking in one line.",
      "description": "Five hero headline+subhead pairs ranked by clarity.",
      "category": "marketing",
      "targetModels": [
        "GPT-4o",
        "Gemini"
      ],
      "tags": [
        "copywriting",
        "landing-page"
      ],
      "isFavorite": false,
      "usageCount": 4,
      "lastUsedAt": "2026-08-26T12:38:34.296Z",
      "createdAt": "2026-08-26T14:38:34.296Z",
      "updatedAt": "2026-08-26T14:38:34.296Z"
    },
    {
      "id": "def5fe13-9a8e-4f72-a0dc-78505c370edf",
      "title": "Regex explainer + rewriter",
      "body": "Explain what this regular expression matches, piece by piece:\n\n/{{regex}}/\n\nThen: (a) write a plain-English one-liner, (b) if it can be simplified without changing behavior, show the simpler version.",
      "description": "Explains a regex in English, then simplifies it.",
      "category": "coding",
      "targetModels": [
        "Claude",
        "GPT-4o"
      ],
      "tags": [
        "explainer",
        "regex"
      ],
      "isFavorite": false,
      "usageCount": 5,
      "lastUsedAt": "2026-08-26T12:38:34.680Z",
      "createdAt": "2026-08-26T14:38:34.680Z",
      "updatedAt": "2026-08-26T14:38:34.680Z"
    },
    {
      "id": "c201efee-d96d-463e-8532-034e323caa2c",
      "title": "Customer interview synthesizer",
      "body": "Synthesize these {{n}} customer interviews about {{topic}}.\n\nOutput:\n1. Themes (name + supporting quotes, cited by interviewee)\n2. Contradictions between interviewees\n3. Jobs-to-be-done statements (\"When ___, I want to ___ so I can ___\")\n4. Top 3 unanswered questions for future interviews\n\nINTERVIEWS:\n{{transcripts}}",
      "description": "Clusters raw interview quotes into themes with evidence.",
      "category": "research",
      "targetModels": [
        "Claude"
      ],
      "tags": [
        "synthesis",
        "user-research"
      ],
      "isFavorite": true,
      "usageCount": 10,
      "lastUsedAt": "2026-08-26T12:38:35.165Z",
      "createdAt": "2026-08-26T14:38:35.165Z",
      "updatedAt": "2026-08-26T14:38:35.165Z"
    },
    {
      "id": "9dacf21e-810e-462c-860c-1fd020805e67",
      "title": "Translation with glossary",
      "body": "Translate the text below from {{source_language}} to {{target_language}}.\n\nGlossary (never translate these terms):\n{{glossary}}\n\nTone: {{tone}}. Preserve formatting and line breaks.",
      "description": "Translates text while respecting a term glossary.",
      "category": "writing",
      "targetModels": [
        "GPT-4o",
        "Gemini"
      ],
      "tags": [
        "localization",
        "translation"
      ],
      "isFavorite": false,
      "usageCount": 2,
      "lastUsedAt": "2026-08-26T12:38:35.647Z",
      "createdAt": "2026-08-26T14:38:35.647Z",
      "updatedAt": "2026-08-26T14:38:35.647Z"
    },
    {
      "id": "30a7577e-0868-4deb-a05b-7bd309d461f3",
      "title": "Regex-free log parser",
      "body": "Parse the {{log_type}} log lines below into a markdown table with columns: {{timestamp}} | level | component | message | trace_id (if present).\n\nDrop health-check noise lines (containing \"healthcheck\" or \"/api/health\"). If a line fails to parse, put it in a second table \"UNPARSED\".\n\nLOGS:\n{{logs}}",
      "description": "Asks the model to parse logs into structured rows.",
      "category": "analysis",
      "targetModels": [
        "GPT-4o"
      ],
      "tags": [
        "logs",
        "parsing"
      ],
      "isFavorite": false,
      "usageCount": 0,
      "lastUsedAt": null,
      "createdAt": "2026-08-26T14:38:36.129Z",
      "updatedAt": "2026-08-26T14:38:36.129Z"
    }
  ],
  "tags": [
    {
      "name": "copywriting",
      "count": 2
    },
    {
      "name": "persona",
      "count": 2
    },
    {
      "name": "ads",
      "count": 1
    },
    {
      "name": "api",
      "count": 1
    },
    {
      "name": "blog",
      "count": 1
    },
    {
      "name": "changelog",
      "count": 1
    },
    {
      "name": "classification",
      "count": 1
    },
    {
      "name": "code-generation",
      "count": 1
    },
    {
      "name": "code-review",
      "count": 1
    },
    {
      "name": "content",
      "count": 1
    },
    {
      "name": "curation",
      "count": 1
    },
    {
      "name": "data",
      "count": 1
    },
    {
      "name": "developer-relations",
      "count": 1
    },
    {
      "name": "ecommerce",
      "count": 1
    },
    {
      "name": "editing",
      "count": 1
    },
    {
      "name": "email",
      "count": 1
    },
    {
      "name": "engineering",
      "count": 1
    },
    {
      "name": "explainer",
      "count": 1
    },
    {
      "name": "extraction",
      "count": 1
    },
    {
      "name": "few-shot",
      "count": 1
    },
    {
      "name": "fiction",
      "count": 1
    },
    {
      "name": "general",
      "count": 1
    },
    {
      "name": "image",
      "count": 1
    },
    {
      "name": "incident",
      "count": 1
    },
    {
      "name": "json",
      "count": 1
    },
    {
      "name": "landing-page",
      "count": 1
    },
    {
      "name": "learning",
      "count": 1
    },
    {
      "name": "localization",
      "count": 1
    },
    {
      "name": "logs",
      "count": 1
    },
    {
      "name": "meetings",
      "count": 1
    },
    {
      "name": "parsing",
      "count": 1
    },
    {
      "name": "plain-language",
      "count": 1
    },
    {
      "name": "productivity",
      "count": 1
    },
    {
      "name": "regex",
      "count": 1
    },
    {
      "name": "research",
      "count": 1
    },
    {
      "name": "review",
      "count": 1
    },
    {
      "name": "sales",
      "count": 1
    },
    {
      "name": "seo",
      "count": 1
    },
    {
      "name": "sql",
      "count": 1
    },
    {
      "name": "sre",
      "count": 1
    },
    {
      "name": "strategy",
      "count": 1
    },
    {
      "name": "summarization",
      "count": 1
    },
    {
      "name": "synthesis",
      "count": 1
    },
    {
      "name": "system-prompt",
      "count": 1
    },
    {
      "name": "testing",
      "count": 1
    },
    {
      "name": "translation",
      "count": 1
    },
    {
      "name": "user-research",
      "count": 1
    }
  ]
}