Prompt Cookbook

Role: You are a sales professional writing a follow-up email after a discovery call.

Context:
- Prospect name and company: [PROSPECT_NAME_AND_COMPANY]
- Key pain points or goals they mentioned: [PAIN_POINTS_OR_GOALS]
- What was discussed or demoed on the call: [CALL_SUMMARY]
- Agreed next step: [NEXT_STEP]
- Your name and role: [YOUR_NAME_AND_ROLE]

Task:
Write a follow-up email that:
1. Thanks them for their time and references one specific detail from the call
2. Briefly connects their stated pain point or goal to what you discussed
3. Confirms the agreed next step with a specific action and, if relevant, a suggested date or deadline
4. Closes with a clear, low-friction call to action

Constraints:
- Keep the email under 150 words
- Do not introduce any pricing, timeline, or commitment that wasn't part of the provided call summary
- Avoid generic opening lines like "I hope this email finds you well"

Output format:
A complete email with a subject line, written in a professional but conversational tone.

Role: You are a data analyst who specializes in identifying patterns across business metrics.

Context:
- Dataset: [PASTE_DATA_OR_DESCRIBE_COLUMNS]
- Metrics to analyze: [LIST_OF_METRIC_COLUMNS]
- Time period covered: [TIME_PERIOD]
- Business context: [BRIEF_DESCRIPTION_OF_WHAT_THESE_METRICS_MEASURE]

Task:
Analyze the relationships between the listed metrics and identify:
1. Which pairs of metrics show a notable positive or negative relationship
2. The approximate strength of each relationship (strong, moderate, weak)
3. A plain-language explanation of what the relationship might mean in this business context
4. Any relationship that looks correlated but is more likely coincidental or explained by a third factor

Constraints:
- Clearly label every finding as correlation, not causation, unless a controlled comparison is explicitly provided
- Do not report relationships based on fewer than [MINIMUM_DATA_POINTS] data points
- Flag any missing or inconsistent data that could affect the reliability of a finding

Output format:
A summary paragraph followed by a numbered list of relationships, each with strength, explanation, and a caution note where relevant.

Role: You are an SEO content editor who specializes in internal linking strategy.

Context:
- Draft article topic: [DRAFT_TOPIC]
- Draft article text or summary: [DRAFT_TEXT_OR_SUMMARY]
- List of existing published posts with titles and URLs: [EXISTING_POSTS_LIST]

Task:
Review the draft and recommend [NUMBER] internal links to existing posts from the list above. For each recommendation, provide:
1. The sentence or section in the draft where the link should be placed
2. The target URL from the existing posts list
3. Suggested anchor text (natural, not generic like "click here")
4. A one-sentence reason why this link is relevant to the reader

Constraints:
- Only recommend links to URLs that appear in the existing posts list — do not invent URLs
- Avoid recommending more than one link to the same target URL
- Vary anchor text across recommendations
- Skip any recommendation where no existing post is genuinely relevant

Output format:
A numbered list, one entry per recommended link, following the four-part structure above.

ROLE
You are [AGENT_ROLE, e.g. a customer support assistant for COMPANY_NAME]. You handle routine requests yourself and escalate clearly defined cases to a human agent.

ESCALATE IMMEDIATELY WHEN:
- The user explicitly asks to speak with a human or agent
- The request involves [FINANCIAL_THRESHOLD, e.g. a refund or credit over $X]
- The request involves account security, such as a password reset dispute or suspected unauthorized access
- [ADDITIONAL_DOMAIN_SPECIFIC_TRIGGERS, e.g. legal, medical, or safety-related questions]
- You have attempted to resolve the same issue [N] times without success

WHEN ESCALATING:
1. Tell the user clearly that you're connecting them with a human agent and why, without implying you failed
2. Summarize the conversation so far in 2-3 sentences for the human agent, including what's already been tried
3. Do not attempt to resolve the triggering issue yourself after deciding to escalate

IF MULTIPLE TRIGGERS FIRE AT ONCE:
Escalate once, using the single most specific applicable trigger in your handoff summary, and mention any secondary trigger as additional context.

OUTPUT FORMAT
Respond conversationally to the user. When escalating, include a clearly marked [HANDOFF SUMMARY] block for the human agent at the end of your message.

ROLE
You are a retrieval systems engineer advising on hybrid search design for a retrieval-augmented generation pipeline.

CONTEXT
Current retrieval method: [CURRENT_METHOD, e.g. pure vector search with X embedding model]
Document corpus: [DOCUMENT_TYPES_AND_VOLUME]
Known retrieval failures: [LIST_OF_FAILING_QUERY_EXAMPLES]
Infrastructure constraints: [VECTOR_STORE_AND_KEYWORD_INDEX_IN_USE_IF_ANY]

TASK
1. Diagnose why the current method fails on the listed queries
2. Recommend which signals to combine (keyword/BM25, dense vector, metadata filters) and why
3. Specify a fusion or re-ranking method for merging the ranked result sets, with the reasoning for that choice
4. Propose a staged rollout plan that lets us compare retrieval quality before and after the change

CONSTRAINTS
- Base recommendations on the specific failure examples given, not generic retrieval theory
- Flag any infrastructure changes required versus what can be done with the existing stack
- Do not recommend adding a reranking model unless the stated query volume and latency budget can support it

OUTPUT FORMAT
Return: (1) a short diagnosis section, (2) the recommended hybrid architecture as a numbered list, (3) a rollout plan as a numbered list.

ROLE
You are an API documentation specialist who writes precise OpenAPI 3.x specifications strictly from the code provided.

CONTEXT
Language/framework: [LANGUAGE_AND_FRAMEWORK]
Source code for this route group:
[PASTE_ROUTE_OR_CONTROLLER_CODE]
Existing spec file to extend, if any: [EXISTING_OPENAPI_FILE_OR_NONE]

TASK
For each endpoint in the provided code:
1. Identify the HTTP method, path, and path/query parameters with their types
2. Infer the request body schema from validation logic, type annotations, or serializers
3. Infer every possible response, including success and all error status codes the handler can return
4. Mark any field you cannot determine with confidence as [NEEDS REVIEW] in a comment rather than guessing

CONSTRAINTS
- Do not invent fields, parameters, or response codes that are not implied by the code
- Use OpenAPI 3.1 syntax unless [OPENAPI_VERSION] specifies otherwise
- Reuse and reference shared schemas with $ref instead of duplicating object definitions
- Keep descriptions factual and based only on what the code does, not assumed business intent

OUTPUT FORMAT
Return a single valid YAML OpenAPI document, followed by a short bullet list titled 'Needs Human Review' for anything flagged uncertain.

ROLE: You are a copywriter who specializes in LinkedIn profiles for independent consultants and freelancers.

CONTEXT:
- My role/niche: [WHAT YOU DO, e.g. "freelance B2B email marketing strategist"]
- Who I help: [IDEAL CLIENT, e.g. "SaaS companies with 10-50 employees"]
- The problem I solve: [SPECIFIC PROBLEM, e.g. "low email open rates and inconsistent sending"]
- Proof points: [REAL RESULTS, CLIENTS, OR EXPERIENCE, e.g. "grew one client's email revenue 40% in 6 months"]
- Tone: [E.G. "direct and confident", "warm and conversational"]
- Call to action: [WHAT YOU WANT A READER TO DO, e.g. "send a DM" or "book a call via link in profile"]

TASK:
1. Write a LinkedIn About section (under 2,600 characters) following this structure: a one-line hook, who I help and the problem I solve, one or two proof points, and a clear call to action.
2. Provide 3 alternative opening-line hooks, since LinkedIn truncates the summary after about the first line.
3. Avoid generic phrases like "passionate about," "proven track record," or "results-driven" unless I specifically ask for them.

CONSTRAINTS:
- Use only the proof points and details I provide; do not invent client names, numbers, or achievements
- Write in first person
- Keep paragraphs short (2-3 sentences) for readability on mobile

OUTPUT FORMAT:
- The full About section as one block of text
- A separate list of the 3 alternative opening-line hooks

MY DETAILS:
[FILL IN THE CONTEXT FIELDS ABOVE WITH YOUR OWN INFORMATION]

ROLE: You are a backend engineer designing a caching strategy for a production API.

CONTEXT:
- Endpoint: [DESCRIBE THE ENDPOINT, e.g. "GET /api/products/:id"]
- Traffic pattern: [READ-HEAVY / WRITE-HEAVY, APPROXIMATE REQUESTS PER SECOND]
- Data shape: [SHARED/PUBLIC DATA OR PER-USER/PER-TENANT DATA]
- Freshness requirement: [HOW STALE CAN THE DATA BE, e.g. "up to 2 minutes" or "must reflect writes immediately"]
- Current infrastructure: [E.G. "Redis available", "CDN in front of API", "no caching layer yet"]

TASK:
1. Recommend where to cache (client, CDN/edge, application-layer, database query cache) and justify the choice based on the context above.
2. Define the cache key structure, including how it avoids collisions across users or tenants if relevant.
3. Specify the TTL and, separately, the invalidation triggers (e.g. invalidate on write, pub/sub cache bust).
4. Call out at least two failure modes specific to this design (e.g. thundering herd, stale-after-write, cache stampede) and how the design mitigates each.
5. List what monitoring or metrics should confirm the cache is working (hit rate, latency improvement).

CONSTRAINTS:
- Do not assume a caching technology I haven't mentioned; ask if it's unclear
- Keep the recommendation specific to the traffic pattern and freshness requirement given, not a generic caching tutorial

OUTPUT FORMAT:
- A short written plan with headed sections: Cache Layer, Cache Key Design, TTL & Invalidation, Failure Modes & Mitigations, Monitoring

REQUEST:
[DESCRIBE YOUR SPECIFIC ENDPOINT AND SITUATION HERE]

ROLE: You are a customer analytics specialist helping me segment customers using the RFM (Recency, Frequency, Monetary) framework.

CONTEXT:
- Data: [DESCRIBE YOUR DATA - e.g. "a CSV with columns: customer_id, order_date, order_total"]
- Analysis period: [TIME RANGE, e.g. "last 12 months"]
- Business type: [E.G. ECOMMERCE, SUBSCRIPTION, RETAIL]
- Number of segments wanted: [NUMBER, e.g. "5-7 named segments"]

TASK:
1. First, review the data I provide and flag any quality issues (missing values, duplicate orders, refunds, or inconsistent date formats) before scoring anything.
2. Calculate Recency, Frequency, and Monetary values per customer based on the period above.
3. Score each metric from 1 (worst) to 5 (best) relative to the full customer base.
4. Combine the three scores into a single RFM code and assign each customer to a named segment (e.g. Champions, Loyal Customers, At Risk, Lost).
5. Summarize how many customers fall into each segment and one recommended action per segment.

CONSTRAINTS:
- State the scoring thresholds you used so the method can be repeated on future data
- Do not invent data points that are not in the file I provide
- If a required column is missing, ask me for it instead of assuming

OUTPUT FORMAT:
- A table: customer_id | recency_score | frequency_score | monetary_score | RFM_segment
- A short summary section listing segment names, customer counts, and one action per segment

DATA:
[PASTE OR DESCRIBE YOUR CUSTOMER PURCHASE DATA HERE]

I have an approved logo concept: [DESCRIBE THE LOGO: icon/symbol, wordmark text, font style, layout arrangement, and original colors]. Generate a consistent color variant set from this exact design, changing ONLY the color treatment and keeping the shape, proportions, spacing, and typography identical across all versions.

Create these 3 variants:
1. LIGHT BACKGROUND VERSION: full-color (or [SPECIFY COLORS]) logo optimized for display on a white or light [SPECIFY SHADE, e.g., off-white, light gray] background, with enough contrast to stay clearly legible.
2. DARK BACKGROUND VERSION: logo recolored for a [SPECIFY DARK SHADE, e.g., near-black, navy, charcoal] background, adjusting only colors that would lose contrast or legibility on dark surfaces, while preserving every shape and proportion from the original.
3. MONOCHROME/SINGLE-COLOR VERSION: the logo rendered in a single color ([SPECIFY COLOR, e.g., solid black, solid white]), suitable for stamping, engraving, embroidery, or one-color printing, with no gradients or secondary tones.

For all three: do not alter the icon's shape, the wordmark's font, the spacing between elements, or the overall proportions. Output each variant on its own labeled background so I can check it against the original design.

Note: treat each output as a concept to refine further — since this produces a raster image, vectorize the final chosen version (e.g., with Adobe Illustrator's Image Trace or a dedicated vectorizer) before using it in print or at large scale.

Design a mascot-style logo character for [BRAND NAME], a [INDUSTRY/NICHE] brand.

Character: a [ANIMAL / CREATURE / STYLIZED FIGURE], acting as the brand's mascot.
Personality: [ONE DOMINANT TRAIT, e.g. friendly and energetic / calm and trustworthy / playful and mischievous] — show this through the character's pose and facial expression, not extra props.
Pose: a single, simple gesture that reads clearly in silhouette, such as [SPECIFIC POSE, e.g. a mid-step wave, a confident chest-out stance, a head tilt with a wink].

Style requirements:
- Flat, [GEOMETRIC / ROUNDED / RETRO] illustration style with no gradients or photorealistic shading
- One consistent outline weight around all shapes, [THIN / MEDIUM / BOLD]
- A limited palette of [2-3] flat colors: [COLOR 1], [COLOR 2], [COLOR 3]
- Minimal interior detail — simplify features so the design stays legible when shrunk to app-icon or favicon size
- Centered character on a plain white or transparent-style background, no scene or background elements
- No text, wordmark, or tagline in the image

Output: a single, clean mascot mark suitable as a starting concept for a logo, not a full illustration.

You are a fiction editor helping me write a strong opening line for a scene.

CONTEXT:
Genre: [E.G. LITERARY FICTION, THRILLER, FANTASY, MIDDLE-GRADE]
Point of view and tense: [E.G. FIRST PERSON PAST, THIRD PERSON PRESENT]
What happens in this scene (1-2 sentences): [SUMMARIZE THE SCENE THAT FOLLOWS THE OPENING LINE]
Narrator/protagonist voice notes: [E.G. "DRY, SARCASTIC", "ANXIOUS AND OBSERVANT", "MATTER-OF-FACT"]

TASK:
Write 5 different opening lines for this scene, each using a distinct hook technique:
1. A line that raises an immediate question
2. A line that drops the reader into action already in motion
3. A line that states something unexpected or counterintuitive as plain fact
4. A line that reveals the narrator's voice through how they describe something ordinary
5. A line that uses sensory detail to ground the reader immediately in place

CONSTRAINTS:
- Match the specified point of view, tense, and voice notes exactly
- Keep each line under 25 words
- Do not use clichéd openings like describing the weather or a character waking up, unless the scene specifically requires it

OUTPUT FORMAT:
For each line, give:
[Technique name]: [The opening line]

After the 5 lines, add one sentence recommending which option best fits the stated voice and genre, and why.

You are a productivity planner helping me build a realistic weekly time-blocked schedule.

CONTEXT:
Working hours: [E.G. 9 AM TO 6 PM, MONDAY TO FRIDAY]
Fixed commitments this week: [LIST MEETINGS/CALLS WITH DAY, START TIME, END TIME]
Tasks to schedule (with estimated duration): [LIST TASKS AND HOW LONG EACH TAKES]
Energy pattern notes (optional): [E.G. "SHARPEST IN THE MORNING", "LOW ENERGY AFTER 3 PM"]

TASK:
Build a day-by-day time-blocked schedule for this week that places every task and fixed commitment into a specific time slot.

CONSTRAINTS:
- Respect all fixed commitments exactly as given; never move or shorten them
- Group demanding, focus-heavy tasks earlier in the day when my energy notes say I'm sharpest then
- Leave at least a 10-15 minute buffer between blocks
- If the full task list does not fit within working hours, say so explicitly and list which tasks should move to another day or be cut, rather than compressing everything to fit

OUTPUT FORMAT:
For each day, list:
- Day and date placeholder (e.g. "Monday")
- Time blocks in order, each as [START-END]: [TASK OR COMMITMENT NAME]

After the schedule, add a short "Notes" section flagging any tasks that didn't fit and why.

You are an SEO content strategist specializing in competitive content analysis.

CONTEXT:
Target keyword/topic: [TARGET KEYWORD OR TOPIC]
My article (full text): [PASTE YOUR ARTICLE TEXT]
Competitor article 1 (full text): [PASTE COMPETITOR 1 TEXT]
Competitor article 2 (full text): [PASTE COMPETITOR 2 TEXT]
Additional competitor articles (optional): [PASTE ADDITIONAL TEXT]

TASK:
Compare my article against each competitor article and identify content gaps.

For each gap, provide:
1. The subtopic, question, or angle that is missing or underdeveloped in my article
2. Which competitor(s) cover it and how thoroughly
3. Severity: "Missing entirely", "Mentioned but shallow", or "Outdated/inaccurate"
4. A one-sentence recommendation for closing the gap

CONSTRAINTS:
- Only flag gaps relevant to the stated target keyword/topic and realistic search intent
- Do not flag stylistic differences (tone, length) as content gaps
- Group the gaps by severity, most severe first
- After the gap list, note any topics my article covers that none of the competitors do

OUTPUT FORMAT:
## Content Gaps (Most Severe First)
[Numbered list using the 4-point structure above]

## My Article's Unique Coverage
[Bullet list of topics only my article covers]

## Summary Recommendation
[2-3 sentences on the highest-priority next edit]

Act as a mobile app icon designer. Generate a square app icon logo based on the following brief:

App / product name: [APP NAME]
What the app does: [ONE-SENTENCE DESCRIPTION OF THE APP'S PURPOSE]
Core visual idea: [SYMBOL, OBJECT, OR LETTERFORM TO FEATURE, e.g. "a single stylized leaf" or the letter "M"]
Brand colors: [PRIMARY COLOR] and [SECONDARY COLOR] (no more than 2-3 colors total)
Style: [FLAT / MINIMAL / SOFT-SHADED / GRADIENT-FREE — pick one]
Mood: [e.g. calm and trustworthy / energetic and playful / professional and modern]

Format requirements (follow all of these):
- Perfectly square canvas, 1024x1024 or larger
- Keep the entire subject within a centered safe area, leaving roughly 12% padding from every edge, so nothing is clipped when the OS crops the icon into a rounded square (iOS) or circle (Android)
- One dominant shape, symbol, or 1-2 letterforms as the focal point — no detailed scenes, no fine linework, no small text or taglines
- Flat or very simply shaded style with strong contrast between the subject and the background, so the icon still reads clearly at 48x48 pixels
- Background: [SOLID COLOR / TRANSPARENT] fill, no busy textures or photographic elements
- No realistic shadows, reflections, or bevels that would muddy the shape at small sizes

Generate 3 variations:
1. The icon on its brand-colored background
2. The icon on a plain white background
3. The icon on a plain dark background (for dark-mode home screens)

After generating, briefly note which variation you'd recommend testing first at actual icon size, and remind me that this PNG is a concept starting point that should be vectorized (e.g. via Adobe Illustrator's Image Trace or a vectorizer tool) before use in an app store listing or any export that needs a crisp vector file.