What These Prompts Do
These are thinking tools. They help you analyze contracts, audit content, decompose problems, surface blind spots, and stress-test plans. They make your judgment sharper. They do not replace it.
What They Don't Do
They don't write your bulletin, your homily, your donor letters, or your social posts. Generative content prompts fail the dignity test the moment a parishioner reads something they believe came from a person and didn't. These prompts are different. The output stays inside your head and your operations. It never reaches the person you are serving without passing through you first.
How to Use
Copy the prompt block. Replace the bracketed sections with your specifics. Paste into Claude or ChatGPT. Iterate. Save the variations that work.
Vendor and Contract Analysis
Vendor Nonsense Detector
Translates vendor marketing into operational reality and surfaces hidden costs.
You are a procurement consultant who has reviewed 500+ software contracts for nonprofits and Catholic institutions. I am evaluating [vendor name] for [purpose]. Their pitch materials are below.
Translate the marketing language into plain English. For each major claim, tell me:
1. What the claim actually means in operational terms
2. What it does NOT include
3. What I should ask to verify it
Then list:
- Hidden costs likely not in the proposal (integration, training, data migration, support tiers, overage fees)
- Lock-in mechanisms (data formats, API access, contract auto-renewal, exit fees)
- 10 questions I should ask before signing
Do not recommend a decision. Surface what I cannot see.
[Paste vendor materials]
Contract Lock-In Finder
Reads a contract and lists every clause that makes leaving harder.
You are a contracts attorney specializing in software and services agreements. Below is a contract I am reviewing.
Identify every clause that affects my ability to leave this vendor, including:
- Auto-renewal terms and notice windows
- Data portability and export rights
- Termination fees and early-exit penalties
- Required notice periods
- Restrictions on switching providers
- Data retention and deletion obligations after termination
- Non-compete or exclusivity clauses
- Pricing escalators after the initial term
For each clause, quote the exact language and explain in plain English what it means.
End with one paragraph on how hard it would be to leave this vendor in year 2, year 3, and year 5.
[Paste contract]
Vendor Comparison Matrix
Compares proposals on the criteria that matter, not the ones vendors emphasize.
You are a procurement consultant. Below are proposals from [number] vendors for [purpose].
Build a comparison matrix covering:
- Total 5-year cost (not just year 1)
- Implementation timeline and resources required from us
- Data ownership and portability terms
- Integration with existing systems
- Support model (response times, channels, costs)
- Training included vs. extra
- Customization scope and cost
- Exit terms
Then identify:
- Which vendor is best on cost
- Which is best on flexibility
- Which is best on integration
- Where each vendor's proposal is vague or evasive
- What I would need to ask each vendor to make a real comparison
Do not recommend a winner. Show me the tradeoffs.
[Paste proposals]
Pre-Signature Question Generator
Generates the questions vendors hate to answer before you sign.
You are a procurement consultant. I am about to sign [type of contract] with [vendor]. The final terms are below.
Generate 20 questions I should ask before signing, organized by:
- Operational reality: what is daily use actually like
- Hidden costs: what is not in the price
- Failure modes: what happens when X breaks
- Exit: what happens if we leave
- References: who has done this and would they do it again
Prioritize questions vendors hate to answer.
[Paste contract or terms]
Decision Support
Strategic Decision Framework Builder
Builds a complete decision framework for any high-stakes call.
You are an organizational strategist working with a Catholic institution. I am facing this decision: [describe].
Build me a decision framework covering:
1. Stakeholder map: who is affected, who has authority, who needs to be consulted
2. Second-order consequences: what changes downstream
3. Reversibility: how hard is it to undo
4. Three strongest arguments for
5. Three strongest arguments against
6. Devil's advocate position from the perspective most likely to disagree
7. Blind spots I might be missing given who I am and where I sit
8. Decision criteria: what would have to be true for this to be the right call
9. What I would need to know that I don't currently know
Do not recommend a decision. Help me think.
Context: [history, constraints, anything relevant]
Resource Allocation Interrogator
Surfaces where stated priorities and actual spending diverge.
You are a financial analyst who has audited budgets for 100+ Catholic dioceses and parishes. I am reviewing [budget area or full operating budget].
Surface:
1. Where stated priorities and actual spending diverge
2. Sacred cows: line items not questioned in 5+ years
3. Hidden subsidies: where one program is quietly funding another
4. Dependencies: which line items lock us into other line items
5. Reallocation scenarios: if we cut X by 20%, where would the impact land
Ask me 5 uncomfortable questions about why we spend what we spend.
Do not recommend cuts. Surface contradictions.
Budget: [paste or describe]
Reversibility Check
Tells you the cost of being wrong before you commit.
You are a risk analyst. I am considering [decision or action]. Tell me:
1. Worst plausible outcome if this goes wrong
2. What it would take to reverse this in 30 days, 6 months, 2 years
3. What becomes harder to change once this is done
4. Who would bear the cost of reversal
5. Signals that would tell me to reverse course, and how early I would see them
6. Pilot or staged versions that would reduce reversibility risk
Do not recommend whether to proceed. Tell me the cost of being wrong.
Pre-Mortem
Imagines the project has failed and works backward to find what we missed.
You are an organizational risk analyst. I am about to begin [project, rollout, or decision]. Imagine it is 18 months from now and the project has clearly failed.
Tell me:
1. The 5 most likely failure modes, ranked by probability
2. For each: early warning signs that would have been visible at month 3
3. Decisions we are making right now that look reasonable but caused the failure
4. What we did not do that we should have done
5. What we are about to dismiss as "fine" that turned out to matter
Do not reassure me. Help me see what I am about to miss.
Project: [describe]
Communication and Content Audit
Communication Problem Diagnostician
Finds the actual problem when messages aren't landing.
You are a communications strategist for nonprofit and Catholic institutions. I have a communication problem: [symptoms — low engagement, repeated questions, etc.].
Before diagnosing, ask me 10 questions covering:
- Audience and who actually receives the message
- Channel and timing
- Format and length
- Stated purpose vs. actual purpose
- Feedback mechanisms
- What the audience does in response
After my answers, give me:
1. Most likely root cause with confidence level
2. Two alternative hypotheses
3. Three test scenarios I could run this month
4. Success metrics that would confirm the new approach is working
Stated problem is rarely the actual problem. Find the actual one.
Content Audit Framework
Tells you what you already have before you create anything new.
You are a content strategist. Below is an inventory of existing content [website pages, bulletins, social posts, internal docs].
Build:
1. Matrix of: piece, last updated, purpose, audience, current performance if known
2. Gap analysis: important topics with no content
3. Refresh priority: highest-stakes outdated content first
4. Cannibalization map: where pieces compete for the same intent
5. Consolidation opportunities: where 5 weak pieces could become 1 strong piece
6. Retirement list: content that exists for institutional reasons but serves no audience
Do not generate new content. Tell me what we already have.
[Paste inventory]
Stakeholder Translation Engine
Tells you what each audience actually needs to hear.
You are a communications strategist. I need to communicate the following to multiple stakeholder groups: [topic, decision, change].
Groups: [pastor, finance council, young families, elderly parishioners, diocese, ministry leaders, etc.]
For each group:
1. What they actually care about (not what they say they care about)
2. Unstated concerns
3. Language that resonates
4. Language that repels
5. Likely objections and how to address them
6. Who in the group influences the rest
Do not write the actual messages. Tell me what each group needs to hear.
Reading Level and Plain Language Check
Shows you what would need to change to make a piece accessible.
You are a plain-language editor. Below is a piece intended for [audience].
Tell me:
1. Current Flesch-Kincaid reading level
2. Sentences over 25 words (list them)
3. Words with simpler equivalents (list pairs)
4. Jargon, acronyms, or insider terms a newcomer would miss
5. Passive voice instances
6. Where the main point is buried
Do not rewrite. Show me what would need to change to bring it to a 6th-grade reading level.
[Paste content]
Operations and Planning
Meeting Agenda Optimizer
Tells you whether the meeting needs to exist before it gets scheduled.
You are an organizational effectiveness consultant who has redesigned 1000+ church and nonprofit meetings. My planned meeting:
- Attendees: [list with roles]
- Duration: [time]
- Stated purpose: [what we are trying to accomplish]
- Topics on current agenda: [list]
Give me:
1. Ruthless prioritization: which topics actually need this group, this duration, in person
2. Pre-work that would shrink the meeting
3. Optimized agenda with explicit decision points (not just discussion)
4. What to cut entirely or move offline
5. Who could be removed from the invite without harm
6. Whether this meeting needs to exist at all
Do not be polite. Tell me where we are wasting time.
Problem Decomposition Specialist
Separates the entangled sub-problems hiding inside one big problem.
You are a strategic consultant. I am facing this problem: [describe as I currently understand it].
Decompose it:
1. Separate the entangled sub-problems (most "single problems" are 3-5 issues)
2. For each sub-problem: actual question, what solving it looks like, how it connects to the others
3. Dependencies: what must be solved first
4. Quick wins (30 days) vs. long plays (6+ months)
5. What is outside my control entirely
6. What I am treating as a constraint that might actually be negotiable
Do not propose solutions. Show me the structure of the problem.
Timeline Stress-Test
Finds where a timeline breaks before reality finds it for you.
You are a project manager who has run 200+ implementations for nonprofit and church institutions. Below is a proposed timeline for [project].
Stress-test it:
1. Every assumption baked into the timeline
2. For each major milestone: most likely cause of delay
3. Critical path: which delays cascade and which are absorbed
4. Single points of failure: what one absence or breakdown stalls the project
5. Buffer reality: actual slack vs. paper slack
6. The 30% scenario: what does this look like if everything takes 30% longer
Do not give me a revised timeline. Tell me where this one breaks.
[Paste timeline]
Process and Knowledge Transfer
Process Documentation Interviewer
Surfaces the tribal knowledge that vanishes when one person leaves.
You are a process consultant. [Person] is [leaving, retiring, training a replacement]. Their core responsibility is [describe].
Interview me with 15-20 questions covering:
- Daily, weekly, monthly tasks
- Decision criteria: when do they say yes, no, escalate, hold
- Relationships: who they call when X breaks, who calls them
- Edge cases and exceptions
- Tribal knowledge: who knows what, where things are, why we do things this way
- Tools, accounts, vendor relationships
After my answers, generate:
1. Decision tree for the role's core decisions
2. Checklist of recurring tasks by frequency
3. Training outline for the replacement
4. Risk assessment: what knowledge is at highest risk of being lost
Most parish processes exist entirely in one person's head. Surface what would vanish.
Organizational Health Diagnostic
Surfaces dysfunction patterns the people inside have stopped noticing.
You are an organizational psychologist with experience in Catholic institutions. I work at [type of organization] and things feel off, but I cannot articulate why.
Ask me 12 diagnostic questions covering:
- Decision-making patterns: who decides, how fast, what gets stuck
- Communication flow: formal vs. informal, where information lives
- Conflict: how it surfaces or doesn't
- Energy: where it accumulates, where it drains
- New ideas: what happens to them
- Departures: who leaves, why, how it's discussed
After my answers:
1. Three dysfunction patterns most likely active
2. Cultural assessment: what the institution actually rewards vs. what it says it rewards
3. Pressure points: where dysfunction shows up first
4. Leverage points: where one change shifts several patterns
5. 12-month roadmap of structural changes (not tactics)
Dysfunctional organizations normalize their dysfunction. Surface what insiders cannot see.
The Self-Check
Dignity Filter Audit
Asks AI to evaluate whether your own use of AI passes the dignity test.
You are an ethicist trained on the Human Test framework: AI must serve human dignity, never substitute for human connection, and pass five obligations (data sovereignty, transparency, reversibility, proportionality, knowing when to stop).
I am about to use AI for [describe specific task or workflow].
Audit my use against the framework:
1. Trajectory: does this move people closer to or further from human connection?
2. Deception: would the affected person feel deceived if they knew AI was involved?
3. Capability: does this make staff more capable, or more dependent?
4. Zone classification: Green (tool), Yellow (assistant), or Red (substitute)?
5. Five obligations, one by one:
- Data sovereignty: where does the data go, who owns it, is it being used to train external models?
- Transparency: are the people affected informed?
- Reversibility: how do we undo this if it harms someone?
- Proportionality: is the AI use proportionate to the problem?
- Knowing when to stop: what is the threshold at which we stop using AI here?
End with one of three verdicts: do not use AI here, use it with these specific conditions, or use it freely.
Be willing to tell me to stop.
Customizing These Prompts
- Be specific. "Warm tone" is vague. "Grateful but not gushy" is clear.
- Provide as much context as you can. AI works better knowing your audience and constraints.
- Set clear constraints. Word counts, format, what NOT to include.
- Request variations. Ask for 3 versions, take what works.
- Adapt the role. The strategist becomes a canon lawyer, liturgist, fundraising consultant, finance analyst. Same prompt, different lens. It shouldn't matter, but it really does change the results you get.
What These Prompts Will Never Do
They will not write content that reaches a person without passing through a human. They will not handle confidential pastoral matters, personnel issues, or anything from the confessional. They will not replace theological discernment, formation, or judgment. The dignity filter holds at every step.
If you sense that a prompt's output makes you more dependent and less capable, stop using it. The point is to think more clearly, not to think less.
