For nonprofit and public sector teamsTemplates

Check every requirement is covered, for nonprofit and public sector teams

Go through a list of requirements one at a time, find where your document answers each one, and get a short list of the gaps to close before you submit, without inventing any missing detail.

This is the version for nonprofit and public sector teams. The steps and checks are the same as in the general template; each prompt carries one added line about your setting, and the examples are from your kind of work.

Sample results: runs and scores on this page are illustrations while testing is set up. How we test

  • Who it is for: Grant writers, program managers and small agency staff who answer a call for proposals, a request for bids or an agency checklist with an application or a bid.
  • Time: 15 minutes

Download it, in the format you use

  • Prompt chain

    Copy each step into ChatGPT or Claude, one after another.

  • Checklist

    Follow the steps and check off the checks by hand.

  • Claude skill

    A folder Claude loads, so it follows the steps by itself.

  • n8n file

    Import into n8n and connect your own AI account.

Using a download:How to use this in n8nHow to install a Claude skill

How it runs

Select a part to see its prompt, input, output and check.

The same job, with and without the template

Sample

One plain prompt

Check whether my application meets all of the requirements and tell me what is missing. (The same requirements and application are pasted below.)

Your application meets most of the requirements.

  1. Responsible person and email: met. Imani Brooks, imani@example.org.
  2. Total amount and budget: met. You ask for 48,000 dollars and list drivers, fuel and refrigerated bags as the costs.
  3. Households served: met. About 120 households in the first year.
  4. Plan for collecting feedback: not mentioned. You should add one.
  5. Two earlier programs with a contact: met. The winter coat drive and the school lunch program are described.
  6. Continuing after the grant ends: met. Six local businesses have pledged to cover the delivery costs.

The only gap is the feedback plan.

Problem:It counts a list of costs as a budget with staff costs shown separately and two earlier programs as met though only one has a contact, so it reports one gap where there are three.

With this template

The same input, run through the steps.

Result: six requirements checked: three covered, two partly covered, one not covered.

Gaps to close

Swipe sideways to see all columns

#Requirement (quote)What is missingWhere to add it
2"state the total amount requested in dollars, with a budget that shows staff costs separately"The total is given, but there is no budget showing staff costs separately.In the sentence that gives the amount asked for
4"include a plan for collecting feedback from participants"The application says nothing about collecting feedback.A new paragraph after the sentence on households served
5"describe two earlier programs, with a contact for each"Two programs are described, but only the winter coat drive has a contact.In the Past work sentence

Covered

  • "name the person responsible for the program and give an email address"
  • "say how many households the program will serve in its first year"
  • "say how the program will continue after the grant ends"
  • Check:Every requirement is quoted word for word: Six requirements quoted word for word from the requirements, each with what the application must show.
  • Check:Every passage is quoted word for word from the document: Six rows: five passages quoted word for word from the application, and one None found (the feedback plan).
  • Check:Each requirement is marked covered, partly covered or not covered: Six judgments, each one of the three allowed words: three covered, two partly covered, one not covered.
  • Check:Every gap row is filled in: Three rows under Gaps to close, each with all three cells filled.
  • Check:Every requirement in the gap list is in the requirements word for word: All three quoted requirements are in the requirements word for word.
  • Check:The counts add up to the judgments: The Result line says six requirements, three covered, two partly covered and one not covered; the table has three rows, one for each requirement that is not fully covered.
  • Check:No gap is filled with a detail nobody gave: No row states or suggests a figure, name, date or method; each says only what is missing and where it belongs.
  • For your read:You decide which gaps matter: Does the missing staff-cost breakdown matter to the grant program that set the requirements?

Runs and scores

Sample

Claude, Sonnet model

October 4, 2026, easy sample

  • Check:Every requirement is quoted word for word

    Three requirements quoted word for word, each with what the application must show.

  • Check:Every passage is quoted word for word from the document

    Three rows, each with a passage quoted word for word from the application.

  • Check:Each requirement is marked covered, partly covered or not covered

    Three judgments, each one of the three allowed words: all three covered.

  • Check:Every gap row is filled in

    One row under Gaps to close, reading None found in every cell.

  • Check:Every requirement in the gap list is in the requirements word for word

    The gap table holds only the None found row, so no quote is out of place.

  • Check:The counts add up to the judgments

    The Result line says three requirements, three covered, none partly covered and none not covered; the gap list is the one None found row.

  • Check:No gap is filled with a detail nobody gave

    No row states or suggests a figure, name, date or method.

  • For your read:You decide which gaps matter

    With no gaps listed, do you still agree each requirement is met in full?

Fixed: Nothing needed fixing: every check passed on the first pass.

Met 8 of 8

1 of these is for your read: the tool hands it to you.

ChatGPT, GPT model

October 5, 2026, tricky sample

  • Check:Every requirement is quoted word for word

    Five requirements quoted word for word, each with what the bid must show.

  • First pass failed:Every passage is quoted word for word from the document

    The first version gave the passage for the staff requirement as "led by Sana Qureshi, with a team of three", which is not in the bid word for word.

  • Check:Each requirement is marked covered, partly covered or not covered

    Five judgments, each one of the three allowed words: two covered, two partly covered, one not covered.

  • Check:Every gap row is filled in

    Three rows under Gaps to close, each with all three cells filled.

  • Check:Every requirement in the gap list is in the requirements word for word

    All three quoted requirements are in the requirements word for word.

  • Check:The counts add up to the judgments

    The Result line says five requirements, two covered, two partly covered and one not covered; the gap list has three rows.

  • Failed:No gap is filled with a detail nobody gave

    The row for the records requirement suggests keeping records for seven years, a period nobody gave.

  • For your read:You decide which gaps matter

    Can you confirm the date the bid will be sent, since the deadline depends on it?

Fixed: The find step ran again once with the rule to quote word for word, and it came back quoting "The work will be led by Sana Qureshi, with two translators and a proofreader", which cleared the first failure. The report step ran again once and the records row still suggested seven years, so the check failed again and the run stopped there and flagged it for you.

Met 7 of 8

  • Not met:No gap is filled with a detail nobody gave: The records row still suggests seven years, a period nobody gave.

1 of these is for your read: the tool hands it to you.

n8n, import format 2.20

October 6, 2026, typical sample

  • Check:Every requirement is quoted word for word

    Six requirements quoted word for word, each with what the application must show.

  • Check:Every passage is quoted word for word from the document

    Six rows: five passages quoted word for word from the application, and one None found.

  • Check:Each requirement is marked covered, partly covered or not covered

    Six judgments, each one of the three allowed words: three covered, two partly covered, one not covered.

  • Check:Every gap row is filled in

    Three rows under Gaps to close, each with all three cells filled.

  • Check:Every requirement in the gap list is in the requirements word for word

    All three quoted requirements are in the requirements word for word.

  • First pass failed:The counts add up to the judgments

    The first Result line said two covered and three partly covered, but the judgments show three covered and two partly covered.

  • Check:No gap is filled with a detail nobody gave

    No row states or suggests a figure, name, date or method.

  • For your read:You decide which gaps matter

    Does the missing staff-cost breakdown matter to the grant program that set the requirements?

Fixed: The report step ran again once with the rule to count from the judgments, and the Result line then read three covered, two partly covered and one not covered.

Met 8 of 8

1 of these is for your read: the tool hands it to you.

The steps, and the check on each

  1. Step 1: List each requirement and what it asks for

    Split the requirements into separate items, quoted word for word, and say what the document must show for each.

    Hands back: A numbered table of the requirements, each quoted and with what the document must show.

    Check:Every requirement is quoted word for word

    A plain rule. If it fails, run the step again (the n8n file does this for you).

  2. Step 2: Find where each one is answered

    For each requirement, quote the words in your document that answer it, or say that none do.

    Hands back: A table that puts a quoted passage from the document beside each requirement, or None found.

    Check:Every passage is quoted word for word from the document

    A plain rule. If it fails, run the step again (the n8n file does this for you).

  3. Step 3: Judge each requirement

    Compare what each requirement asks for with the passage found, and say what is missing.

    Hands back: A table that marks each requirement covered, partly covered or not covered and names what is missing.

    Check:Each requirement is marked covered, partly covered or not covered

    A plain rule. If it fails, run the step again (the n8n file does this for you).

  4. Step 4: Write the gap report

    Turn the judgments into a count, a list of gaps with where each should be added, and a list of what is covered.

    Hands back: A report with a count line, a table of gaps and a list of the requirements that are covered.

    Check:Every gap row is filled in

    A plain rule. If it fails, run the step again (the n8n file does this for you).

After the last step

  • Check:Every requirement in the gap list is in the requirements word for word

    A plain rule. If it fails, run the step again (the n8n file does this for you).

  • Check:The counts add up to the judgments

    A second look by the AI. If it fails, run the step again (the n8n file does this for you).

  • Check:No gap is filled with a detail nobody gave

    A second look by the AI. If it fails, run the step again (the n8n file does this for you).

  • Check:You decide which gaps matter

    You check this one. If it fails, it is flagged for you.

What you need first

  • The requirements(needed)

    Paste the requirements exactly as they were given to you, in the order they appear.

  • Your document(needed)

    Paste the full text of the document that has to meet the requirements, as it stands now.

Ways to adapt it

  • This template checks your document against the list as written. It cannot tell whether a requirement is reasonable, and it does not know requirements that were given to you out loud.
  • If the requirements come with weights or must-have marks, paste them into the requirements text and add a column for them in the list step.
  • If a requirement depends on something outside the texts, such as a date or a page limit, the template marks it partly covered and says what is needed, so check that one yourself.
  • To change the coverage words, change them in the judge step's rules and in its check, both.
  • For a long list of requirements, run the template on one section of the list at a time.

Change any step or check in the Editor, then download it again.

Words that fit nonprofit and public sector teams

  • Request for proposals, call for proposals and notice of funding opportunity all name the document that sets the requirements; an agency buying a service may call it a solicitation or a request for bids.
  • Applicant, bidder and respondent are the same writer; use the word the requirements use when you read the gap list.
  • A requirement tied to a date, such as a submission deadline, stays partly covered until you confirm the date yourself.
Try it with a sampleSample

Six requirements and a short application; three are fully met, two are met only in part (a budget with staff costs shown separately, a contact for each earlier program) and one is not answered at all.

The requirements
Applications must name the person responsible for the program and give an email address. They must state the total amount requested in dollars, with a budget that shows staff costs separately. They must say how many households the program will serve in its first year. They must include a plan for collecting feedback from participants. They must describe two earlier programs, with a contact for each. They must say how the program will continue after the grant ends.
Your document
Application: Neighborhood Pantry Delivery Program. Our program is led by Imani Brooks (imani@example.org). We are asking for 48,000 dollars, which covers drivers, fuel and refrigerated bags. In its first year the program will serve about 120 homebound households across three zip codes. Past work: we ran a winter coat drive for the town library in 2024 (contact: Hector Salas, hector@example.org) and a school lunch program last spring. After the grant ends, six local businesses have pledged to cover the delivery costs.

The scoring guide

Sample

One criterion for each check, marked met or not met. (The guides in Quality rate a whole piece of work on three levels.) Under each one is how the three sample runs scored.

Download the scoring guide (Markdown)

  • Every requirement is quoted word for word

    A plain rule

    Meets:Every requirement in the table is found word for word in the requirements.
    Does not meet:A requirement has been reworded, or is not in the requirements word for word.
    Claude:Met
    ChatGPT:Met
    n8n:Met
  • Every passage is quoted word for word from the document

    A plain rule

    Meets:Every passage is found word for word in the document, or the row says None found.
    Does not meet:A passage has been reworded, or is not found word for word in the document.
    Claude:Met
    ChatGPT:Met
    n8n:Met
  • Each requirement is marked covered, partly covered or not covered

    A plain rule

    Meets:Every requirement has exactly one of covered, partly covered or not covered.
    Does not meet:A row has another word, such as mostly covered or met.
    Claude:Met
    ChatGPT:Met
    n8n:Met
  • Every gap row is filled in

    A plain rule

    Meets:Every row under Gaps to close has the requirement, what is missing and where to add it filled in.
    Does not meet:A row under Gaps to close has a blank or a dash in one of its three cells.
    Claude:Met
    ChatGPT:Met
    n8n:Met
  • Every requirement in the gap list is in the requirements word for word

    A plain rule

    Meets:Every requirement quoted in the gap list is found word for word in the requirements.
    Does not meet:A quoted requirement has been reworded or is not in the requirements.
    Claude:Met
    ChatGPT:Met
    n8n:Met
  • The counts add up to the judgments

    A second look by the AI

    Meets:The numbers on the Result line match the judgments, and the gap list has one row for each requirement that is not fully covered.
    Does not meet:A count on the Result line is wrong, or a requirement that is not fully covered has no row in the gap list, or a covered one has.
    Claude:Met
    ChatGPT:Met
    n8n:Met
  • No gap is filled with a detail nobody gave

    A second look by the AI

    Meets:No row in the gap list states or suggests a figure, name, date or promise that is not in the requirements or the document.
    Does not meet:A row fills a gap with a detail nobody gave, such as a guarantee length or a price.
    Claude:Met
    ChatGPT:Not met
    n8n:Met
  • You decide which gaps matter

    You check this one

    Meets:You read each gap, decided whether it matters to the person who set the requirements, and added the detail yourself from facts you hold.
    Does not meet:A gap is ignored without a decision, or a detail is added that you cannot back.
    Claude:For your read
    ChatGPT:For your read
    n8n:For your read

Claude: met 8 of 8 (including 1 for your read)

ChatGPT: met 7 of 8 (including 1 for your read)

n8n: met 8 of 8 (including 1 for your read)

Which one fits?

Review

Review

Write

Last run
October 6, 2026: n8n, import format 2.20 (sample run)
Tools and versions
Claude, Sonnet model; ChatGPT, GPT model; n8n, import format 2.20
What was checked
The files, against each tool’s import format, and 3 sample runs, each check’s finding recorded.
Version
1.0
What changed
First release.

Report a problem with this templateHow we test