Resources / Prompt Library / HR & Corporate
For HR, Leadership, Logistics, Client Services & IT
Strategic AI for the Workplace
29 advanced prompts built for corporate teams, each labeled with the prompting technique it demonstrates: chain of thought, tree of thought, few-shot, self-consistency, and more.
29prompts
6role groups
7techniques taught
The Techniques Behind These Prompts
Every prompt below is labeled with the advanced technique it uses, so you learn the method while you use the prompt.
| Chain of Thought | Asks the AI to reason step-by-step before answering. More accurate because the AI shows its work instead of jumping to conclusions. |
| Tree of Thought | Explores multiple approaches before selecting the best one. Ideal when there is no single right answer. |
| Role + Context | Sets up a specific professional identity, situation, and constraints. The more context, the more usable the output. |
| Few-Shot | Provides examples of good output first. Teaches the AI your preferred format and quality standard. |
| Self-Consistency | Generates multiple independent answers and identifies where they agree. Reduces hallucination for high-stakes decisions. |
| Structured Output | Specifies the exact format, sections, and structure. Eliminates reformatting work. |
| Meta-Prompting | Uses AI to generate better prompts, for when you are not sure how to ask for what you need. |
Human Resources
6 prompts90-Day Onboarding Plan Builder
Role + ContextBest tool: CoPilot (integrates with your Microsoft 365 environment)
You are a senior HR manager at a [mid-size logistics/manufacturing/aviation maintenance] company with [number] employees. You are onboarding a new [role title] who will report to [manager title] in the [department] team.
Company context:
- We use [list key software: TMS, ERP, fleet management, SAP, etc.]
- Our onboarding has historically taken [X weeks] before new hires are fully productive
- Key compliance requirements: [DOT regulations, safety certifications, OSHA, industry-specific]
Create a detailed 90-day onboarding plan with the following structure:
WEEK 1: Orientation and setup
- Day-by-day schedule
- Who is responsible for each item (HR, direct manager, IT, assigned buddy)
- Systems access and training needed
- Required paperwork and compliance items
DAYS 8-30: Foundation building
- Key milestones the new hire should reach
- Training modules to complete with deadlines
- Check-in schedule with manager
DAYS 31-60: Growing independence
- Tasks the new hire should own independently
- Cross-team introductions and collaboration goals
- Mid-point performance conversation template
DAYS 61-90: Full integration
- Performance benchmarks for the role
- KPI targets
- 90-day review conversation guide
Format as a table with columns: Date/Timeframe | Task | Owner | Completion Criteria
Employee Integration Risk Assessment
Chain of ThoughtBest tool: ChatGPT or Claude
I'm an HR manager onboarding a new [role title] into our [department] at a [industry] company. I need to identify potential integration risks before they become problems.
Think through this step by step:
Step 1: Based on the role of [role title] in [industry], what are the top 5 common reasons new hires in this type of role fail to integrate within the first 90 days? For each reason, explain WHY it happens, not just what it is.
Step 2: For each risk you identified, what are the early warning signs that would appear in weeks 1-4? Be specific about observable behaviors, not vague descriptions.
Step 3: For each early warning sign, what is the specific intervention the direct manager or HR should take? Include the exact conversation to have and when to have it.
Step 4: Now synthesize this into a one-page "Integration Risk Checklist" that a hiring manager can use during weekly check-ins for the first 90 days. Format as a checklist with Yes/No checkboxes and "If No, do this" action items.
Step 5: Finally, based on your analysis, what is the single highest-impact action our HR team can take in the first week to prevent the most common integration failure?
Hiring Strategy for Hard-to-Fill Roles
Tree of ThoughtBest tool: Claude (handles the multi-path analysis well)
We are a [industry] company trying to fill a [role title] position. We've had the role open for [X weeks/months] with limited qualified applicants. Our location is [city/region] and our budget for the role is [range].
I want you to explore three completely different hiring strategies. For each strategy, think it through independently as if the other two don't exist.
STRATEGY A -- Traditional Recruitment (Optimized)
Think through: What is failing in our current approach? What specific changes to job posting language, channel selection, and screening process would increase qualified applicants? What does the optimized version look like step by step?
STRATEGY B -- Build Instead of Buy
Think through: What if we hired someone with 70% of the qualifications and trained them up? What would that training plan look like? What is the cost of training vs. the cost of the vacancy remaining open? What is the realistic timeline to full productivity?
STRATEGY C -- Restructure the Role
Think through: What if the reason we can't fill this role is that we've designed it wrong? Could this be split into two roles? Could some responsibilities be automated with AI? Could we redesign it as a remote or hybrid position to widen the talent pool?
After exploring all three:
- Compare them on: time to fill, total cost (including vacancy cost), quality of outcome at 12 months, and risk
- Recommend which strategy to pursue first and why
- Provide a decision framework I can present to leadership
Performance Review Comment Generator
Few-ShotBest tool: CoPilot (paste directly into your review system)
I need to write performance review comments for team members. Here are examples of the quality and tone I want:
EXAMPLE 1 (Exceeds Expectations):
Input: "Sarah consistently meets deadlines and helped onboard three new team members"
Output: "Sarah has demonstrated exceptional reliability in meeting project deadlines throughout this review period. Beyond her core responsibilities, she took initiative to mentor three new team members during their onboarding, developing informal training materials that the team continues to use. Her contributions have measurably strengthened team capacity and reduced the typical ramp-up time for new hires."
EXAMPLE 2 (Meets Expectations with Growth Area):
Input: "Mike does solid work on reports but needs to communicate delays earlier"
Output: "Mike consistently produces thorough, accurate reports that meet our quality standards. His analytical skills are a reliable asset to the team. An area for growth is proactive communication when timelines shift. Strengthening his ability to flag potential delays early will build greater confidence among stakeholders who depend on his deliverables."
Now write a performance review comment for this input. Match the tone, specificity, and balanced approach shown in the examples above.
Input: "[Paste your rough notes about the employee here]"
Rating: [Exceeds Expectations / Meets Expectations / Needs Improvement]
Training Needs Analysis
Structured OutputBest tool: Any
I manage HR for a [industry] company with [number] employees across these departments: [list departments]. We are implementing [new technology/process/system] and need a training needs analysis.
Provide your analysis in EXACTLY this structure:
SECTION 1: SKILLS GAP MATRIX
Create a table with columns: Department | Current Skill Level (1-5) | Required Skill Level (1-5) | Gap | Priority (Critical/High/Medium/Low)
SECTION 2: TRAINING PLAN BY PRIORITY
For each Critical and High priority gap:
- Specific training needed (not generic "provide training")
- Recommended format (live workshop, self-paced, on-the-job, vendor certification)
- Estimated time investment per employee
- Internal vs external training recommendation with reasoning
SECTION 3: IMPLEMENTATION TIMELINE
Provide a 12-week rollout schedule. Sequence training so that foundational skills come before advanced ones. Never schedule more than 4 hours of training per employee per week.
SECTION 4: SUCCESS METRICS
For each training module, define: what competency looks like, how to measure it, and the timeline for employees to demonstrate it.
SECTION 5: BUDGET ESTIMATE
Break down by: internal labor costs, external vendor costs, materials, and lost productivity during training.
HR Process Automation Finder
Meta-PromptBest tool: ChatGPT or Claude
I'm an HR professional at a [industry] company. I want to find which of my daily tasks are best suited for AI assistance, but I'm not sure I'm thinking about it the right way.
Here is a list of tasks I do in a typical week:
[Paste your actual weekly task list here, be as specific as possible]
For each task I listed:
1. Score it on the 4R Test: Repetitive (1-5), Restructuring (1-5), Reviewable (1-5), Risk-appropriate (1-5)
2. If the total score is 15 or higher, write me a specific, ready-to-use prompt I could paste into an AI tool to do that task
3. For each prompt you write, label which prompting technique it uses and which AI tool would work best
4. If the total score is below 15, briefly explain why this task should stay human-led
Finally, estimate how many hours per week I would save if I used AI for all the tasks that scored 15 or above.
Leadership & Executive
5 promptsStrategic Resource Allocation Under Constraints
Tree of ThoughtBest tool: Claude (excels at multi-path complex analysis)
I'm a [VP/Director/Leader] at a [industry] company. My team of [number] people handles [list key functions]. Our workload has increased [X]% but headcount is frozen. I need to decide where to focus our limited resources.
Current priorities competing for attention:
1. [Priority 1 - describe]
2. [Priority 2 - describe]
3. [Priority 3 - describe]
4. [Priority 4 - describe]
Explore three different resource allocation strategies. For each, think through it completely and independently:
APPROACH A -- Ruthless Prioritization
If we could only do TWO of these four priorities well, which two create the most organizational value? What do we explicitly deprioritize and what is the real cost of that decision? Walk through the reasoning step by step.
APPROACH B -- AI-Augmented Capacity
For each of the four priorities, what specific tasks could be automated or AI-assisted to reduce the human hours required? Be concrete: which tasks, which tools, and what is the realistic time savings? Could AI free up enough capacity to address all four?
APPROACH C -- Structural Redesign
What if the problem is not resources but structure? Could responsibilities be redistributed, outsourced, or consolidated? Think through which functions could be combined, which could be delegated to a less senior role, and which require redesign.
After exploring all three:
- Which approach yields the best outcome at 90 days? At 12 months?
- What is the risk profile of each?
- Draft a one-page recommendation I can bring to my executive team, including the trade-offs I'm asking them to accept.
Data-Driven Decision Brief
Chain of ThoughtBest tool: ChatGPT or Claude
I need to make a decision about [describe the decision]. I have the following data and context:
[Paste relevant data, metrics, observations, or background information]
Walk me through this decision step by step:
Step 1: Summarize the key data points. What do the numbers actually tell us? Separate facts from assumptions. If any data is ambiguous or incomplete, flag it explicitly.
Step 2: Identify the decision options. What are the realistic choices available? Don't just list two polar opposites. Include any middle-ground or phased approaches.
Step 3: For each option, analyze:
- What does success look like if this works?
- What is the worst realistic outcome?
- What would need to be true for this to be the right choice?
- Who are the stakeholders affected and what do they care about most?
Step 4: What information am I missing that would make this decision easier? What could I look up, ask someone, or test before committing?
Step 5: Given the analysis above, what is your recommendation and why? Be direct. Don't hedge with "it depends" unless you specify exactly what it depends on.
Format the final output as a one-page decision brief I can share with my leadership team, with a clear recommendation at the top and supporting analysis below.
IT Governance and AI Policy Advisor
Self-ConsistencyBest tool: Claude
I lead a team at a [industry] company. Employees are adopting AI tools (CoPilot, ChatGPT, Gemini, Claude) without formal governance. I need to create a practical AI governance framework, but I want to make sure my recommendations are solid.
I want you to approach this problem three separate times, as if you were three different advisors:
ADVISOR 1 -- The Risk-First CISO
Design an AI governance framework that prioritizes security, data protection, and compliance. What policies, controls, and oversight structures would you recommend?
ADVISOR 2 -- The Innovation-First CTO
Design an AI governance framework that prioritizes adoption speed, employee empowerment, and competitive advantage. What policies and structures enable rapid, safe experimentation?
ADVISOR 3 -- The Operations-First COO
Design an AI governance framework that prioritizes practical implementation, minimal bureaucracy, and measurable ROI. What is the simplest framework that still manages risk?
Now compare all three frameworks:
- Where do all three advisors AGREE? (These are your non-negotiable policies.)
- Where do they DISAGREE? (These are your strategic choices -- present the trade-offs.)
- Synthesize a unified framework that takes the strongest element from each perspective.
Deliver the final framework as a document I can present to our executive team, with an implementation timeline.
Executive Communication Drafter
Role + ContextBest tool: CoPilot (if sending via Outlook/Teams)
You are a communication strategist helping a [VP/Director] draft a message to [audience: all-hands, board, direct reports, cross-functional leaders].
Context:
- Topic: [what the communication is about]
- Key decision or announcement: [the core message]
- Audience concerns: [what they are worried about or will ask]
- Tone required: [transparent but confident / urgent but calm / celebratory / empathetic]
- Constraints: [keep under X words, avoid jargon, must include specific call-to-action]
Before you write, think through:
1. What is the ONE thing the audience should remember after reading this?
2. What question will they immediately ask, and how should I preemptively address it?
3. What is the emotional state of the audience right now, and how does that affect word choice?
Now write the communication. Open with the most important information. Address the likely concern in the middle. Close with a clear next step and who to contact with questions.
After the draft, provide:
- A one-sentence summary I can use as the subject line or Slack preview
- The two sentences most likely to be misinterpreted, with suggested revisions
Weekly Operations Dashboard Summary
Structured OutputBest tool: CoPilot or ChatGPT
Here are my raw notes, metrics, and observations from this week:
[Paste your rough notes, data points, bullet points -- as messy as they are]
Transform this into an executive operations summary in EXACTLY this format:
HEADLINE: One sentence summarizing the most important thing leadership needs to know this week.
GREEN (On Track):
- [Item]: [One-sentence status with metric if available]
(List 3-5 items)
YELLOW (Needs Attention):
- [Item]: [What is happening] → [What action is being taken] → [When it will be resolved]
(List any items that are at risk)
RED (Escalation Required):
- [Item]: [What is happening] → [What decision is needed from leadership] → [Deadline for that decision]
(List any items requiring leadership intervention)
KEY METRICS THIS WEEK:
| Metric | This Week | Last Week | Trend |
(Only include metrics I provided data for. Do not fabricate numbers.)
NEXT WEEK PRIORITIES:
1. [Priority with owner]
2. [Priority with owner]
3. [Priority with owner]
Keep the total summary under 400 words. Flag anything where the data I gave you is incomplete or unclear.
Logistics & Fleet Operations
5 promptsWork Order Generator from Field Notes
Role + ContextBest tool: Claude (handles messy, long input precisely)
You are a fleet maintenance coordinator at a [regional trucking company / carrier / logistics firm]. Your job is to convert raw field technician notes into professional maintenance work orders for our Fleet Management System (FMS).
Company context:
- We operate a fleet of [number] [vehicle types]
- Our FMS requires structured work orders with specific fields
- DOT compliance requires documentation of all safety-related repairs
- Our service priority levels are: Critical (vehicle off-road, safety issue), High (vehicle operational but degraded), Medium (scheduled maintenance due), Low (cosmetic or convenience)
Convert these raw field notes into a formatted work order:
[PASTE RAW FIELD NOTES HERE]
Required output format:
WORK ORDER #: [Generate: WO-YYYY-MMDD-XXX]
VEHICLE ID:
DATE REPORTED:
REPORTED BY:
CURRENT LOCATION:
VEHICLE STATUS: [In Service / Out of Service / Limited Service]
ISSUES (grouped by priority):
For each issue:
- Priority: [Critical/High/Medium/Low] with justification
- Description: (professional language, complete sentences)
- Probable Cause:
- Required Parts: (with part numbers if inferable)
- Estimated Labor Hours:
- DOT Compliance Flag: [Yes/No -- Yes if this issue affects roadworthiness]
ESTIMATED TOTAL DOWNTIME:
DEADLINE FOR NEXT UPDATE:
NEXT ACTION:
FLAG any item where the field notes are ambiguous and a follow-up question to the technician is needed.
Route and Load Optimization Analyst
Chain of ThoughtBest tool: ChatGPT or Claude
I manage logistics for a [carrier/distributor/freight company]. I need to optimize a delivery schedule. Here are the parameters:
Vehicles available: [number and types]
Deliveries to make: [list destinations, approximate load sizes, time windows]
Starting location: [hub/warehouse address or city]
Constraints: [driver hours limits, vehicle weight limits, customer delivery windows, toll avoidance preferences, fuel stop requirements]
Think through this step by step:
Step 1: Group the deliveries by geographic proximity. Which ones naturally cluster together into routes?
Step 2: For each cluster, estimate the total load weight/volume. Flag any cluster that exceeds a single vehicle's capacity and split it.
Step 3: Sequence the stops within each route to minimize total drive time. Consider delivery time windows -- which stops MUST happen at specific times and how does that constrain the sequence?
Step 4: Assign vehicles to routes based on load requirements and vehicle capabilities. Flag any mismatch.
Step 5: Check each route against driver hours of service regulations. Will any route require a mandatory break? If so, where is the best break point?
Step 6: Identify the single biggest inefficiency in this plan and propose an alternative that fixes it.
Present the final plan as a table: Route # | Vehicle | Driver | Stop Sequence | Estimated Departure | Estimated Arrival at Each Stop | Total Miles | Total Hours
Repetitive Process Elimination
Tree of ThoughtBest tool: Claude
I run [fleet operations / logistics / warehouse] for [company type]. My team spends too much time on repetitive documentation and manual processes. Here is a list of our most time-consuming repetitive tasks:
1. [Task 1 -- describe what it is, who does it, how often, how long it takes]
2. [Task 2]
3. [Task 3]
4. [Task 4]
5. [Task 5]
For each task, I want you to independently evaluate three elimination strategies:
PATH A -- Full AI Automation
Can this task be fully handled by AI with only final human review? What would the AI input and output look like? What is the realistic error rate and how would errors be caught?
PATH B -- AI-Assisted (Human + AI)
Can AI do 70% of this task and a human finishes it? What does the handoff look like? Where exactly does the human take over?
PATH C -- Process Redesign
Does this task even need to exist in its current form? Could we eliminate it entirely, combine it with another task, or change when/how it happens to reduce the burden?
After evaluating all three paths for each task, create a prioritized implementation plan:
- Rank all tasks by: (time saved per week) x (ease of implementation)
- Recommend which to tackle first, second, third
- For the top priority, write me a complete AI prompt I can start using immediately
Incident and Safety Report Standardizer
Few-ShotBest tool: Any
I need to convert informal incident descriptions into standardized safety reports. Here are examples of the format I need:
EXAMPLE 1:
Input: "Driver Mike clipped a bollard backing into dock 7 Tuesday afternoon. Minor scrape on rear bumper passenger side. No injuries. Dock camera caught it."
Output:
INCIDENT REPORT #: IR-2026-0318-001
Date/Time: Tuesday, [date], approximately 1200-1700 hours
Location: Dock 7, [Facility Name]
Vehicle: [To be filled -- unit number needed]
Driver: Mike [Last name needed]
Incident Type: Property Damage -- Minor
Description: During a backing maneuver at Dock 7, the vehicle made contact with a fixed bollard, resulting in a minor scrape on the rear bumper, passenger side. No structural damage to the bollard or the vehicle's operational components was observed.
Injuries: None reported
Witnesses: None reported; dock security camera footage available
Immediate Action Taken: [To be completed]
Follow-up Required: Review camera footage, complete vehicle damage assessment, driver coaching if applicable
DOT Reportable: No
EXAMPLE 2:
Input: "Forklift 12 dropped a pallet in aisle C, some product damaged. Nobody hurt. Tuesday morning around 10."
Output:
INCIDENT REPORT #: IR-2026-0318-002
Date/Time: Tuesday, [date], approximately 1000 hours
Location: Warehouse Aisle C, [Facility Name]
Equipment: Forklift Unit 12
Operator: [Name needed]
Incident Type: Product Damage -- Equipment Related
Description: During standard material handling operations in Aisle C, Forklift Unit 12 dropped a loaded pallet, resulting in damage to product. Quantity and value of damaged goods to be assessed.
Injuries: None reported
Witnesses: [To be identified]
Immediate Action Taken: [To be completed]
Follow-up Required: Product damage assessment, forklift inspection, operator equipment check review
OSHA Reportable: No
Now convert this incident description into the same format:
[PASTE YOUR INFORMAL INCIDENT DESCRIPTION HERE]
Flag any fields where the original description is missing required information, and phrase the flag as a specific question to ask the person who reported it.
Fleet Status Report Builder
Structured OutputBest tool: CoPilot or ChatGPT
Here are my raw fleet notes for this [day/week]:
[Paste all notes: vehicle statuses, maintenance updates, driver issues, delivery completions, delays, fuel data, compliance items -- as messy as you have them]
Organize into this EXACT format:
FLEET STATUS REPORT -- [Date]
SUMMARY: [2-3 sentences: total vehicles in service, total out of service, key highlight, key concern]
VEHICLES IN SERVICE: [count] of [total fleet]
VEHICLES OUT OF SERVICE: [count]
| Vehicle ID | Issue | Est. Return Date | Status |
DELIVERY PERFORMANCE:
| Metric | Today/This Week | Target | Status |
| On-time delivery rate | | | |
| Loads completed | | | |
| Loads pending | | | |
MAINTENANCE SUMMARY:
- Completed this period: [list]
- In progress: [list with expected completion]
- Scheduled upcoming: [list with dates]
COMPLIANCE ITEMS:
- [Any DOT, safety, or regulatory items due or overdue]
ESCALATIONS:
- [Anything requiring management attention, with recommended action]
Only include data I provided. Mark any field where my notes are incomplete with "[DATA NEEDED: specific question]".
Banking & Client Services
4 promptsClient Onboarding Process Optimizer
Chain of ThoughtBest tool: Claude
I work in [banking/financial services/client services]. Our client onboarding process is a bottleneck. Here is how it currently works:
[Describe your current process: steps involved, who does what, what information is collected, typical timeline from first contact to fully onboarded client, where delays happen]
Analyze this process step by step:
Step 1: Map the current process into a numbered sequence. For each step, note: who does it, how long it takes, what information it requires, and what the output is.
Step 2: Identify the bottlenecks. Where do things stall? Is it waiting for client information? Internal approvals? Manual data entry? Compliance review? Rank bottlenecks by how many days they add to the total timeline.
Step 3: For each bottleneck, determine if AI could reduce the delay. Be specific: what would the AI do, what would the input be, what would the output be, and how much time would it save? Only recommend AI where it genuinely helps -- not every bottleneck is an AI problem.
Step 4: For the bottlenecks that are NOT AI problems, what process changes would help? (Resequencing steps, parallel processing, better templates, earlier information gathering)
Step 5: Design the optimized onboarding process. Show the new sequence with estimated time per step. Calculate the total timeline reduction.
Step 6: Write me three AI prompts I can start using immediately to address the top three AI-solvable bottlenecks you identified.
Client Data Gathering Questionnaire Builder
Role + ContextBest tool: ChatGPT
You are a client relationship manager at a [bank/financial services firm/professional services firm]. You need to gather comprehensive information from a new [client type: commercial client, individual account holder, business loan applicant, etc.] efficiently.
The challenge: Clients find long questionnaires tedious, and incomplete submissions create back-and-forth that delays onboarding by [X days/weeks].
Context:
- Information we need: [list all required data fields]
- Regulatory requirements: [KYC, AML, any industry-specific compliance]
- Client type: [size, sophistication level, typical concerns]
- Channel: [email, portal, in-person meeting, phone]
Create a client information gathering approach that:
1. Groups related questions together logically (not by our internal department structure)
2. Starts with the easiest questions to build momentum
3. Explains WHY we need sensitive information (reduces pushback)
4. Flags which items are required vs. optional
5. Includes pre-fill suggestions where we can infer information from public sources
6. Provides a "what to have ready" checklist the client can review BEFORE the main questionnaire
Also provide:
- A short, warm introductory message to send with the questionnaire
- A follow-up template for incomplete submissions that is professional but creates urgency
- A checklist for our team to validate completeness before moving to the next onboarding stage
Client Risk Scoring Framework
Self-ConsistencyBest tool: Claude
I work in [banking/financial services]. I need to develop a framework for assessing [client risk / credit risk / operational risk] for [client type]. This framework will be used by [who will use it] and needs to be both rigorous and practical.
I want you to generate three independent risk scoring frameworks, each using a different methodology:
FRAMEWORK 1 -- Quantitative Scoring Model
Develop a points-based scoring system using measurable data inputs. Define the data fields, point values, thresholds, and how to calculate the composite score. Show the math.
FRAMEWORK 2 -- Qualitative Assessment Matrix
Develop a criteria-based framework using professional judgment categories. Define the assessment dimensions, rating scales, and how to synthesize multiple dimensions into an overall rating.
FRAMEWORK 3 -- Hybrid Approach
Combine quantitative and qualitative elements. Define which factors are best measured with data and which require judgment. Show how the two types of input are weighted and combined.
Now compare the three:
- Which factors appeared in ALL three frameworks? (These are your core risk indicators.)
- Where do they produce different risk ratings for the same client? (These are your judgment calls.)
- Which framework is most practical for a team of [X] to implement immediately?
Recommend the final framework with a rationale. Provide a one-page scoring template my team can start using this week.
IMPORTANT: This framework is a starting tool for human decision-makers, not an automated decision system. All final risk assessments require human review and professional judgment.
Client Communication Standardizer
Structured OutputBest tool: CoPilot (integrates with Outlook)
I need to send a [type of communication: status update / document request / meeting follow-up / issue notification] to a client.
Client context:
- Client name: [name]
- Relationship: [new client / long-standing / high-value / sensitive situation]
- Communication history: [last contact was about X on Y date]
- Tone required: [formal / professional-warm / urgent-but-calm]
Information to convey:
[Paste your rough notes about what needs to be communicated]
Write the communication in this EXACT structure:
SUBJECT LINE: [Clear, specific, action-oriented]
OPENING: [One sentence -- personal reference or context setter, not generic greeting]
CORE MESSAGE: [The main point -- front-loaded, not buried]
DETAILS/NEXT STEPS: [Numbered list of specific actions, owners, and deadlines]
CLOSING: [Clear call-to-action -- what you need from them and by when]
Keep the total under [150/250/400] words. Do not use: "I hope this email finds you well," "per our conversation," "please do not hesitate," or "at your earliest convenience." Use direct, professional language.
IT & Systems
4 promptsTechnical Debt Prioritization
Chain of ThoughtBest tool: Claude (handles complex technical analysis)
I'm [a developer/engineering lead/IT manager] building systems under tight deadlines. We're accumulating technical debt and need to make strategic decisions about what to address now vs. later.
Here is our current technical debt inventory:
[List each item: what it is, where it lives in the codebase/infrastructure, when it was introduced, what risk it poses, and what happens if we ignore it for 6 more months]
Think through this systematically:
Step 1: Categorize each debt item by type:
- Safety debt (will cause outages or data loss if not addressed)
- Velocity debt (slows down development but doesn't cause failures)
- Scalability debt (fine now but breaks at higher volume)
- Maintenance debt (makes everything harder but nothing is broken)
Step 2: For each item, calculate an urgency score based on:
- Probability of causing a problem in the next 90 days (1-5)
- Severity if the problem occurs (1-5)
- Cost to fix now vs. cost to fix after it causes a problem (ratio)
- Dependencies: does fixing this unblock other improvements? (Yes/No)
Step 3: Sequence the work. Given that my team has [X] available hours per sprint for debt reduction, what order should we tackle these in? Show your reasoning.
Step 4: For each of the top 3 priorities, estimate: hours to fix, risk of the fix introducing new issues, and the test plan to validate the fix.
Step 5: Draft a one-page summary I can show to [leadership/stakeholders] explaining why we're dedicating [X]% of our capacity to technical debt reduction and what the cost of NOT doing it would be.
System Architecture Decision
Tree of ThoughtBest tool: Claude
I'm building [describe system or feature] with these requirements:
- Must handle [volume/scale requirements]
- Integrates with [list existing systems: TMS, ERP, FMS, CRM, etc.]
- Team size: [number] developers
- Timeline: [deadline]
- Constraints: [budget, technology restrictions, compliance requirements]
I want you to evaluate three architectural approaches independently:
ARCHITECTURE A: [Option 1 -- e.g., Monolithic / Build Custom / Cloud-Native]
Think through: How would this be built? What are the components? How does it integrate with existing systems? What is the development timeline? What are the operational requirements?
ARCHITECTURE B: [Option 2 -- e.g., Microservices / Buy and Customize / Hybrid]
Think through the same questions independently.
ARCHITECTURE C: [Option 3 -- e.g., Serverless / Integrate Existing Tools / AI-Assisted]
Think through the same questions independently.
Now compare all three on:
| Criteria | Architecture A | Architecture B | Architecture C |
| Time to MVP | | | |
| Total cost (12 months) | | | |
| Maintenance burden | | | |
| Scalability ceiling | | | |
| Team skill requirements | | | |
| Integration complexity | | | |
| Technical debt risk | | | |
Recommend one approach. Explain what you would need to see change about the requirements to recommend a different one.
CI/CD Pipeline Documentation Generator
Role + ContextBest tool: Claude or ChatGPT
You are a senior DevOps engineer writing documentation for a team that includes developers with varying levels of infrastructure experience. The documentation needs to be complete enough that a developer with no DevOps background can follow it without asking for help.
Document our CI/CD pipeline:
Pipeline overview:
[Describe your pipeline: source control, build steps, test stages, deployment targets, environments]
Tools and services:
[List: GitHub/GitLab, Jenkins/GitHub Actions/Azure DevOps, Docker, Kubernetes, AWS/Azure/GCP, monitoring tools, etc.]
Generate documentation that includes:
1. PIPELINE ARCHITECTURE DIAGRAM (describe as a text-based flow diagram)
Show: trigger → build → test → staging → production
Include: what happens at each stage, what can fail, and what failure looks like
2. SETUP GUIDE FOR NEW DEVELOPERS
Step-by-step: environment setup, required access and credentials (without exposing actual secrets), how to run the pipeline locally, how to verify it works
3. COMMON OPERATIONS RUNBOOK
For each operation (deploy, rollback, hotfix, environment promotion):
- When to do it
- Exact commands or steps
- What to verify after
- Who to notify
- What to do if it fails
4. TROUBLESHOOTING GUIDE
The 10 most common pipeline failures, their symptoms, and step-by-step resolution
5. ARCHITECTURE DECISION RECORDS
For each major tooling choice, document: what we chose, what we considered, why we chose this option, and when to reconsider
AI Integration Opportunity Scanner
Meta-PromptBest tool: ChatGPT or Claude
I'm a technology leader at a [industry] company. We have data flowing through these systems:
[List your systems: TMS, FMS, ERP, CRM, HR systems, telematics platforms, etc.]
Currently, humans bridge these systems in these ways:
[Describe manual processes: exporting reports from one system, reformatting data, entering data into another system, comparing outputs across platforms, generating summaries]
Act as an AI integration consultant. For each manual bridge I described:
1. Assess: Is this an AI opportunity, an API integration opportunity, or a process elimination opportunity? Explain why.
2. For AI opportunities: Write me a specific, ready-to-use prompt that would automate or assist with this task. Label the prompting technique used.
3. For API integration opportunities: Describe what the integration would look like, what data flows where, and whether existing tools (Zapier, Power Automate, custom scripts) could handle it.
4. For process elimination: Explain why this process may not need to exist and what would replace it.
Finally, rank all opportunities by: (time saved per week) × (ease of implementation) ÷ (risk if AI makes an error). Give me a prioritized action plan for the next 90 days.
Cross-Functional (Any Role)
5 promptsMeeting Notes to Action Items
Chain of ThoughtBest tool: CoPilot (especially for Teams meetings) or any tool
Here are my rough notes from a [team meeting / client call / strategy session / project standup]:
[Paste your raw notes -- as messy as they are]
Process these step by step:
Step 1: Identify every decision that was made in this meeting. List each one clearly, even if it was stated informally.
Step 2: Identify every action item. For each one, determine: who is responsible (if stated or implied), what specifically they need to do, and the deadline (if stated or implied). If the deadline was not stated, flag it as "DEADLINE NEEDED."
Step 3: Identify any open questions that were raised but not resolved. For each, note who needs to answer it and by when.
Step 4: Identify any risks or concerns that were mentioned, even casually. Things people said like "I'm worried about..." or "that might be a problem" or "we should keep an eye on..."
Step 5: Compile into this format:
MEETING SUMMARY (2-3 sentences)
DECISIONS MADE:
1. [Decision] -- made by [who]
ACTION ITEMS:
| # | Action | Owner | Deadline | Status |
OPEN QUESTIONS:
1. [Question] -- needs answer from [who] by [when]
RISKS/CONCERNS FLAGGED:
1. [Concern] -- raised by [who]
If any of my notes are ambiguous, tell me what you interpreted and flag it for my review.
Email Response Generator
Role + ContextBest tool: CoPilot (in Outlook) or any tool
You are a [role] at a [industry] company. You need to respond to the following email:
[Paste the email you received]
Context for my response:
- My relationship with this person: [manager / client / vendor / colleague / new contact]
- What I want to accomplish: [agree / decline / negotiate / request more info / set a boundary / escalate]
- Constraints: [timeline, budget, authority level, political considerations]
- Tone: [direct / diplomatic / warm / formal]
Before writing the response, think through:
1. What does this person actually need from me? (Not just what they asked for -- what is the underlying need?)
2. What is the best outcome for both of us?
3. Is there anything I should NOT put in writing?
Write the response. Keep it under [100/200/300] words. Lead with my answer or decision, not background context. End with a clear next step.
Problem Diagnosis When You're Stuck
Tree of ThoughtBest tool: ChatGPT or Claude
I'm dealing with a problem at work and I'm stuck. Here's the situation:
[Describe the problem in as much detail as you can. Include: what's happening, when it started, who's involved, what you've already tried, and what makes it difficult]
I want you to diagnose this from three completely different angles:
LENS 1 -- It's a People Problem
Assume the root cause is about people: communication, incentives, skills, relationships, or organizational dynamics. What does the diagnosis look like through this lens? What would the solution be?
LENS 2 -- It's a Process Problem
Assume the root cause is about how work flows: missing steps, unclear ownership, bad handoffs, wrong sequence, or missing feedback loops. What does the diagnosis look like? What would the solution be?
LENS 3 -- It's a Tools/Systems Problem
Assume the root cause is about the tools, technology, or systems: the wrong tool for the job, missing automation, information silos, or outdated systems. What does the diagnosis look like? What would the solution be?
Now step back:
- Which lens best explains the evidence I described?
- Is it possible it's a combination? If so, which lens addresses the root cause vs. the symptoms?
- What is the single most impactful action I could take this week, regardless of which lens is correct?
Professional Document Tone Converter
Few-ShotBest tool: Any
I need to adjust the tone of a document. Here are examples of the transformation I want:
EXAMPLE -- Casual to Professional:
Before: "Hey team, just wanted to flag that the Q3 numbers are looking rough. We're gonna need to hustle on the pipeline if we want to hit target. Let me know if you need anything."
After: "Team, Q3 revenue is currently tracking 12% below target. Closing the gap will require focused pipeline acceleration over the next six weeks. I have identified three specific actions I recommend we take. Please review and share your input by Friday."
EXAMPLE -- Technical to Executive:
Before: "The API latency spikes are being caused by an N+1 query problem in the order processing microservice. We need to refactor the data access layer to use batch queries instead of individual calls per order line item."
After: "Our ordering system is experiencing slowdowns that affect customer experience during peak hours. The engineering team has identified the root cause and has a fix ready to implement. The fix requires three days of development and will reduce processing delays by approximately 80%. I recommend we prioritize this in the current sprint."
Now apply the same transformation to this text:
[Paste your text]
Transform it from [casual/technical/internal/verbose] to [professional/executive/client-facing/concise]. Maintain all factual content. Do not add information I did not provide.
Build Your Own Custom Prompt
Meta-PromptBest tool: Any
I want to create a reusable AI prompt for a task I do regularly. Here's the task:
TASK: [Describe what you do]
FREQUENCY: [How often -- daily, weekly, monthly]
INPUT: [What raw information you start with]
OUTPUT: [What the finished product should look like]
QUALITY STANDARD: [How you know the output is good enough]
COMMON MISTAKES: [What usually goes wrong when this task is done poorly]
Based on this information:
1. Write me a production-ready prompt I can save and reuse. Include:
- A role/context setup
- Clear input instructions with placeholders I'll fill in each time
- Specific output format requirements
- Quality checks built into the prompt itself
2. Label which prompting technique(s) you used and why you chose them for this specific task.
3. Tell me which AI tool (CoPilot, ChatGPT, or Claude) would work best for this prompt and why.
4. Give me one tip for making the output even better over time (how to iterate and improve the prompt based on results).
This library came from a corporate AI strategy workshop.
Eclatech trains HR, operations, and leadership teams on advanced prompting and practical AI workflows: workforce efficiency without the buzzwords.
Book a Corporate Training