Promts

tags: AI Prompts

Tired of AI giving you mediocre outputs because it picked the easiest path? This “task decomposer“ prompt from Excellent AI Prompts forces the model to propose multiple solution methods, compare them side-by-side, and justify its choice before generating anything. 

Here's how it works: Instead of letting the AI jump straight to execution, you make it act as a “systems designer“ that must first propose two different approaches. For each method, it lists 5 specific steps, estimates time required, and identifies the biggest risk. Then it creates a comparison table showing strength, risk, speed, and quality for each approach. Only after choosing one method with a clear rationale does it execute.

Tidy a message: “Rewrite this to be clear and concise. Keep my intent and suggest one alternate subject line: [paste draft].”

1. **Context first.** Start with: “Here’s the context: [paste it].”

2. **Then the task.** “Your task: [what you want], prioritized for [who/what matters most].”

3. **Then constraints + role.** “Act like [role]. Keep it under [length], give [N] options, match this structure: …”

4. **Force questions.** “Before you start, review everything and ask any clarifying questions you need. Number them and make them yes/no where possible.”

5. **Add a correction loop.** After the first answer: “Now critique your response for missing context, shaky assumptions, or fluff, and rewrite it with those fixes.”

Before you start, ask me up to 7 targeted questions that would materially improve the answer. If any info is missing and you must proceed, state your assumptions clearly.

The “Give me options, not one answer” PROMPT

Generate 10 distinct approaches to solve [problem]. 
For each: give a one-line summary, best use case, and one downside. 
Then recommend the top 2 based on [my priority: speed / cost / quality / risk].

“Grounded answer only” Prompt 

Use only the information I provide below. If something isn’t supported, say “I don’t have enough info” and list what you’d need. 
Info: [paste] 
Question: [ask]

“Turn this mess into structure” Prompt 

Organize the following into a structured [doc type: brief, PRD, meeting notes, SOP]. 
Keep all facts, remove repetition, and flag open questions at the end. 
Notes: [paste]

"Before starting [task], work in phases:

1. Ask clarifying questions about [request/context]

2. Propose your approach for [how to do it]

3. Review your initial plan to ensure it aligns with [my goals]

4. Based on your review, write a final plan showing: [what you'll do], [steps], [potential risks]

Keep it scannable but detailed. Wait for my approval before executing."

I want to [TASK] so that [SUCCESS CRITERIA].

First, read these files completely before responding:

1. [filename.md] — [what it contains]

2. [filename.md] — [what it contains]

3. [filename.md] — [what it contains]

DO NOT start executing yet. Instead, ask me clarifying questions (use ‘AskUserQuestion’ tool) so we can refine the approach together step by step.

Only begin work once we’ve aligned.

Your AI is a yes-man. Here's how to fix it in 30 seconds.

Go to **Settings → User Preferences** (Claude) or **Custom Instructions** (ChatGPT) and add:

- Be anti-sycophantic - don’t fold arguments just because I push back

- Stop excessive validation - challenge my reasoning instead

- Avoid flattery that feels like unnecessary praise

- Don’t anthropomorphize yourself

Push back if my assumptions are wrong" or "Tell me what you're uncertain about.

What's the single smartest and most radically innovative addition you could make to this project at this point?

"Give me three different perspectives on [your question]. For each perspective, explain what evidence supports it and what evidence contradicts it. Then tell me which perspective has the strongest support and why."

"Now argue against your own conclusion. What did you miss?"

"Give me one small thing to notice today."

"Share one interesting assumption I might hold without realizing."

"Suggest an overlooked detail worth noticing on my commute."

"Offer a single question that could change my usual perspective." 

"Before you begin, outline your step-by-step plan for completing this task. Wait for my approval or edits before proceeding."

You are a talent branding expert who specializes in writing job descriptions that attract high-quality candidates. I'm going to give you a rough or existing job description, and you will rewrite it to be compelling, clear, and candidate-focused.

Here is the job description I want you to improve:
[PASTE JOB DESCRIPTION HERE]

Rewrite it following these rules:
1. Open with a 2-3 sentence "why this role matters" hook — make the candidate feel the impact they'll have, not just the tasks they'll do
2. Replace vague phrases like "fast-paced environment" or "self-starter" with specific, concrete descriptions of how the team works and what success looks like in 90 days
3. Restructure the requirements into two sections: "Must Have" (true blockers only, max 5) and "Nice to Have" (everything else)
4. Add a "What You'll Actually Do in Your First 30 Days" section with 3-4 specific bullets
5. Close with a 2-sentence company culture statement that sounds human, not corporate
6. Flag any language that may unintentionally deter qualified candidates (e.g., overly gendered words, unnecessary degree requirements)

After the rewrite, give me a 3-bullet summary of the biggest changes you made and why.

I'm going to ask you the same question three times with different framings. For each, give your honest analysis. Then compare all three answers and flag any contradictions.

Framing 1 (supportive): "I'm excited about [topic]. What are the biggest benefits?"
Framing 2 (skeptical): "I'm worried about [topic]. What are the biggest risks?"
Framing 3 (neutral): "Give me a balanced analysis of [topic], including both benefits and risks with evidence for each."

After all three, tell me: did your core conclusions change based on my framing? If so, which answer is closest to your actual assessment?

You are helping me with [specific project/task]. Here's what you need to know:

About me: [Your role, expertise level, industry]

The project: [What you're working on and why]

What good looks like: [Examples, tone, format, length]

Constraints: [What to avoid, word limits, style rules]

Success criteria: [How I'll judge if this is useful]

With this context, please [your actual request].

You are my personal daily briefing agent. Every morning, compile:

1. TOP 3 PRIORITIES from my notes below (rank by deadline, then impact)
2. CALENDAR OVERVIEW: What's on my schedule today, what needs prep
3. INBOX TRIAGE: Flag anything that looks urgent or time-sensitive
4. INDUSTRY PULSE: 3 things I should know about [your industry] today

Here are my inputs:
- Calendar: [paste or describe today's schedule]
- Notes/tasks: [paste your to-do list or project notes]
- Emails to triage: [paste subject lines or summaries]

Format as a scannable briefing I can read in 2 minutes. Bold the one thing I absolutely cannot miss today.

Ask: "Reverse-engineer this conversation into a skill using your skill creator skill I can call anytime." If it doesn’t give you a doc you can “install”, it didn’t work right; see below for more.

If you want to force your AI like Claude to make a new skill or update an old one that’s easy to add to your skills library, make sure you say “Make this a skill / Update this skill with your skill creator skill (scripts.package_skill) to give me a one-click executable to copy to my skills library.” If you don’t use those magic words, it doesn’t work right every time; but every time I request it exactly like this, it works!

I'm going to share [a decision / an idea / a draft]. Your job is to be my devil's advocate. 

Rules:
1. Do NOT validate my idea first. Skip the compliments entirely.
2. List the 3 strongest arguments AGAINST what I'm proposing.
3. Identify the assumption I'm most likely wrong about.
4. Tell me what someone who disagrees with me would say, and why they might be right.
5. Only AFTER doing all of that, tell me what's genuinely strong about it.

Here's what I need feedback on: [paste your thing]

Review our entire conversation history and create a comprehensive personal context document I can take to any AI assistant. Organize it into these sections:

1.PREFERENCES: How I like responses formatted, my communication style preferences, topics I care about
2. CONTEXT: My job, projects, goals, and recurring themes from our conversations
3. KEY DECISIONS: Important choices or conclusions we've reached together
4. ONGOING WORK: Any active projects, drafts, or threads I'd want to continue elsewhere

Format as clean markdown I can copy-paste into a new AI assistant's memory or system prompt.

You are a senior procurement analyst. I'm going to paste a vendor proposal or contract below. Analyze it and return:

**1. Deal Summary (3 bullets max)**
- What they're offering, the price, and the commitment length.

**2. Red Flags**
- List any auto-renewal clauses, hidden fees, vague deliverables, one-sided termination terms, or liability gaps. Quote the exact language that concerns you.

**3. Missing Items**
- What's NOT in this proposal that should be? (e.g., SLAs, data ownership, exit clauses, performance benchmarks, IP rights)

**4. Negotiation Leverage**
- Give me 3 specific counter-points I can bring to the negotiation table, based on what you found.

**5. Verdict**
- Score this deal 1-10 on value-for-money and 1-10 on risk. One sentence recommendation: sign, negotiate, or walk away.

Here's the document:
[PASTE PROPOSAL OR CONTRACT TEXT]

You are a senior contract analyst with 20 years of experience reviewing business agreements.

I will paste a contract below. Your job is to break it down so a non-lawyer business owner can understand every section.

For each clause or section:
1. Give it a plain-English name (e.g., "What You're Paying" instead of "Compensation Terms")
2. Summarize what it actually says in 1-2 simple sentences
3. Rate its favorability:  Standard /  Worth Noting /  Unusual
4. Flag any vague language that could be interpreted against me

At the end, provide:
- A "TL;DR" summary of the entire contract in 5 bullet points
- The 3 most important things I need to understand before signing
- Any sections that are missing but should be there

Here is the contract:
[PASTE YOUR FULL CONTRACT TEXT HERE]

You are a risk management attorney who specializes in protecting businesses from bad contracts.

Using this contract analysis:
[PASTE OUTPUT FROM PROMPT #1]

And the original contract:
[PASTE ORIGINAL CONTRACT]

Conduct a full risk assessment. For each risk found:
1. Name the risk in plain English
2. Quote the exact clause that creates it
3. Explain the worst-case scenario if triggered
4. Rate severity:  Low /  Medium /  High /  Critical
5. Estimate potential financial exposure (range)

Specifically look for:
- Unlimited liability or uncapped damages
- Auto-renewal clauses and cancellation windows
- Non-compete or exclusivity restrictions
- IP ownership transfers or licenses
- Indemnification obligations (who pays if something goes wrong)
- Termination penalties or early exit fees
- Change-of-terms clauses (can they modify the deal unilaterally?)
- Jurisdiction and dispute resolution (where would you get sued?)
- Data ownership and confidentiality traps

Output a ranked risk table from highest to lowest severity. Then give me a "Risk Score" for the overall contract: Low / Medium / High / Walk Away.

You are a contract compliance specialist who helps businesses understand their obligations before signing.

Using this contract analysis and risk assessment:
[PASTE OUTPUT FROM PROMPT #1 AND PROMPT #2]

Create a complete Obligation Map for my side of this agreement:

1. PAYMENT OBLIGATIONS
- What I pay, when I pay, how I pay
- Late payment penalties
- Price escalation clauses

2. PERFORMANCE OBLIGATIONS
- What I must deliver or do
- Quality standards or SLAs I must meet
- Reporting or notification requirements

3. TIMELINE OBLIGATIONS
- All deadlines, milestones, and notice periods
- Auto-renewal dates and cancellation windows
- Response time requirements

4. RESTRICTION OBLIGATIONS
- What I cannot do during the contract
- Non-compete, exclusivity, or non-solicitation terms
- Confidentiality requirements and duration

5. POST-TERMINATION OBLIGATIONS
- What survives after the contract ends
- Return of materials or data requirements
- Ongoing confidentiality or non-compete periods

For each obligation:
- Quote the source clause
- Rate difficulty: Easy / Manageable / Burdensome / Unreasonable
- Flag if it's one-sided (they don't have the same obligation)

End with a calendar view: list every deadline and obligation in chronological order.

I have a recurring task at work: [describe the task].
I currently do it [how often] and it takes about [time].
Build me a simple tool or workflow that automates this.
Walk me through setup step by step, assuming I have zero coding experience.
If it requires any tools or accounts, tell me which ones and how to set them up.

You are an operations expert who specializes in converting messy, real-world business processes into clean, repeatable Standard Operating Procedures (SOPs).

I'm going to describe how we currently handle [PROCESS NAME, e.g., "onboarding a new client," "processing a refund," "publishing a blog post"]. It may be disorganized, incomplete, or inconsistent. Your job is to extract the real workflow and turn it into a structured SOP my team can actually use.

Here's how we currently do it (raw description):
[PASTE YOUR DESCRIPTION HERE — brain dump, Slack thread, email chain, voice memo transcript, or a rough paragraph. The messier the better.]

Based on this, produce:

1. Process Title & Owner — A clear name for this SOP and who owns it.

2. Purpose (2-3 sentences) — Why this process exists and what a successful outcome looks like.

3. When to Use This SOP — The specific trigger or condition that starts this process.

4. Tools & Resources Required — Every tool, document, login, or template needed before starting.

5. Step-by-Step Instructions — Number every step. Write each one as a clear action verb ("Open X," "Send Y," "Check that Z is complete"). Flag decision points with IF/THEN logic (e.g., "If the client hasn't responded in 48 hours, escalate to [ROLE].").

6. Common Mistakes to Avoid — 3-5 specific failure modes or shortcuts that break this process, based on what I described.

7. Definition of Done — A checklist of 3-5 items that confirm this process was completed correctly.

8. Review Cadence — Recommend how often this SOP should be reviewed and updated (monthly, quarterly, etc.) based on how likely this process is to change.

Format the SOP in clean markdown so it can be dropped directly into Notion, Confluence, or Google Docs. Use headers, numbered lists, and bold text for scannability. Do not invent steps I didn't describe — if something is unclear, flag it with [UNCLEAR — CONFIRM WITH TEAM] so I know what to verify.

You are my executive assistant working inside Claude Cowork. You have access to my Gmail through the connector.

Your task: run a complete inbox triage for me right now.

Step 1. Open Gmail and list every unread email from the last 24 hours (or the oldest 50 unread if I have a backlog).

Step 2. For each email, extract:
- Sender name and domain
- Subject line
- One-sentence summary of what they want
- Urgency score (1 to 5, where 5 = needs response today)
- Importance score (1 to 5, where 5 = direct business impact)

Step 3. Sort by (urgency + importance) descending. Output as a clean table.

Step 4. For any email with urgency of 4 or higher, draft a short professional reply in my voice. My voice is: [DESCRIBE YOUR TONE, e.g. "friendly, direct, short sentences, never formal"]. Save each draft inside Gmail so I only need to review and send.

Step 5. For emails with urgency + importance of 3 or lower, suggest which ones I can safely archive. Do not archive anything yourself. Only suggest.

Step 6. End with a 3-bullet summary:
- Top 3 priorities from my inbox today
- Total drafts you created
- Anything I should know about before I open Gmail myself

Do not send any email. Only draft. I review everything.

You are my chief of staff working inside Claude Cowork. You have access to my Google Calendar and Gmail connectors.

Using the inbox priorities you surfaced in Prompt #1 as context, plan my upcoming week.

Step 1. Open Google Calendar and pull every event from tomorrow through next Friday. Include meeting title, attendees, duration, and any attached agenda or doc link.

Step 2. For each meeting:
- Classify it (client, internal, 1:1, external, admin)
- Flag any with no agenda or unclear purpose
- Pull the most recent 3 emails from each external attendee so I walk in with context
- Write a 3-bullet prep note: the goal, the key questions to ask, the outcome I want

Step 3. Find my deep work blocks. A deep work block is any uninterrupted 90-minute gap on the calendar. List every one of them.

Step 4. Match each deep work block to a top priority from my inbox triage. Propose exactly what I should do in each block.

Step 5. Flag red zones: back-to-back meetings with no buffer, double bookings, any day with more than 5 meetings.

Step 6. Output in this exact format:

## Week Overview
[week dates, total meetings, total deep work hours]

## Priority Meetings (with prep notes)
[list with prep bullets]

## Deep Work Blocks
[list with assigned priorities]

## Red Flags
[conflicts, no-agenda meetings, overloaded days]

Do not modify anything in my calendar. Just read and report.

I want you to select a concept at about the graduate-student level from the field of [YOUR FIELD HERE]. Then indirectly explain this concept completely by writing a fable. Structure it so that only toward the very end do readers 
gradually realize what the concept actually is. After the story, add a section that clearly articulates the concept you just conveyed.

You are my strategic chief of staff. I will paste raw inputs from my past week: completed tasks, dropped balls, meeting notes, key metrics, and any open decisions. Your job is to produce a focused 7-day plan.

Do these in order:

1. Summarize last week in 5 bullets: what got done, what slipped, what surprised me, what pattern repeats, what should I stop doing.
2. Identify the single most important outcome I should drive next week. Justify why in one sentence tied to revenue, retention, or risk.
3. Break that outcome into 3 concrete deliverables, each with: owner (me or delegate), Monday start state, Friday done state, and the one thing that could block it.
4. List 3 things I should explicitly NOT do next week, with the cost of saying yes to each.
5. End with a 5-question Friday self-review I can run to grade the week.

Be direct. No fluff. Push back if my inputs are vague.

Inputs:
[paste your week here]

Negotiation Prep → Walk into any high-stakes conversation with leverage, scripts, and fallback positions ready.

You are my negotiation strategist. I have an upcoming negotiation and I need you to prepare me thoroughly so I walk in with leverage, clarity, and a plan.

Here's the situation:
- What I'm negotiating: [describe the deal — e.g., SaaS contract renewal, partnership revenue share, salary, client scope, supplier pricing]
- Who I'm negotiating with: [their role, company, what you know about them]
- My ideal outcome: [the win you actually want]
- My walk-away point: [the line you won't cross]
- What I think they want: [their likely priorities]
- Constraints / context: [timeline, history, dependencies, anything they could exploit]

I want you to deliver:

1. **Power Map** — Who has more leverage right now and why? Identify 2-3 things I'm underestimating about my own position, and 2-3 things they likely have on me.

2. **Their Likely Opening Move** — Predict the first offer or position they'll lead with, and the reasoning behind it. Then give me the exact phrase I should use to respond without anchoring against myself.

3. **Three Concession Trades** — List three things I can give up that cost me little but feel valuable to them, paired with three things I should ask for in exchange.

4. **The Script** — Write the actual opening 3 sentences I should say, plus 5 specific objection-handling responses to the pushback I'm most likely to get.

5. **The Walk-Away Signal** — Give me the exact language to use if we hit my walk-away point, framed so I keep the door open without losing the leverage of being willing to leave.

6. **One Question I'm Not Asking** — Name the single most important question I haven't thought to ask, and explain why it matters.

Be direct. No hedging, no "it depends." Assume I'm sharp but emotionally invested, so flag where my judgment might be off.

You are a contract risk analyst advising a business owner who is NOT a lawyer. I'm about to sign the agreement below. Review it and protect me.

CONTRACT:
"""
[PASTE THE FULL CONTRACT OR CLAUSE HERE]
"""

CONTEXT:
- My role in this deal: [e.g. the customer / the vendor / the contractor / the hiring company]
- What I'm paying or being paid: [amount + cadence]
- Deal size and how much this relationship matters to me: [small/medium/critical]

Do the following:

1. PLAIN-ENGLISH SUMMARY - In 5 bullets, tell me what I'm actually agreeing to, as if explaining to a smart friend.

2. RED FLAGS - List every clause that could hurt me, ranked by severity (high / medium / minor). For each: quote the exact clause text, explain the real-world risk in one sentence, and estimate the worst-case cost or consequence. Pay special attention to: auto-renewal and notice windows, liability caps and indemnification, IP ownership, termination rights, exclusivity/non-compete, payment terms and late fees, and anything that's one-sided.

3. WHAT'S MISSING - Note any standard protection I'd expect that ISN'T here (e.g. liability cap, termination for convenience, confidentiality).

4. REDLINES - For each high and medium item, give me the exact replacement language I should ask for, plus a one-line negotiation script to send the other side.

5. WALK-AWAY TEST - In 2 sentences: is anything in here bad enough that I should not sign without a real lawyer? Be direct.

Do not give generic legal disclaimers beyond one short line. Be specific to MY contract.

Decision Pre-Mortem → Pressure-test a big call before you commit, not after it blows up.

Prompt:

You are a sharp, skeptical operating partner with a track record of stopping
bad decisions before they cost real money. I am about to make the following
decision:

DECISION: [describe the decision in 2-3 sentences]
CONTEXT: [why now, what I expect to gain, what it costs, who it affects]
CONSTRAINTS: [budget, timeline, team, anything fixed]

Run a structured pre-mortem in five steps:

1. FAILURE SIMULATION. Assume it is 6 months from now and this decision has
   clearly failed. Write the 5 most likely stories of how it went wrong,
   ranked by probability. Be specific and concrete, not generic.

2. HIDDEN ASSUMPTIONS. List every assumption I am making that, if false,
   breaks the decision. Flag the 3 most dangerous and tell me how to test
   each one this week for under 100 dollars or a single conversation.

3. REVERSIBILITY CHECK. Rate this decision from 1 (fully reversible) to 5
   (one-way door). If it is a 4 or 5, tell me the smallest possible version
   I could test first to buy back optionality.

4. WHAT WOULD CHANGE MY MIND. Name the single piece of evidence that should
   make me NOT do this, and whether I already have a way to find it.

5. VERDICT. Give me a clear recommendation: proceed, proceed-but-shrink,
   gather-more-data, or kill. One paragraph of reasoning. No hedging.

Ask me up to 3 clarifying questions first if the decision or context is
too thin to analyze well.