Turn a confusing form into a conversation.

What the AI will do

• Read the form and turn its questions into a manageable conversation.
• For a one-time form, search available information for draft answers, ask about what is missing, and prepare the completed form or an answer sheet.
• For repeat use, build and check a conversational helper: an app, interactive artifact, or complete reusable prompt paired with the form.

What you can provide

• Share the form in whatever format you have: a link, PDF, document, or image. The AI will help if it cannot read it.
• For your own form, answer questions the AI cannot resolve from available information.

Where you'll take part

• Choose whether to complete the form once or make a reusable helper. For a helper, confirm its users, intended output, and format before building.
• Confirm or correct every answer the AI supplies before it becomes final, individually or in small related groups. Each person using the helper does the same.
• Review the completed form or helper. Submission, signatures, and final declarations remain your decision.

What you'll get

A completed form or organized answer sheet for final review, or a checked reusable helper that lets other people produce their own answers.

Activity context

A confusing form can become a straightforward conversation. Have AI find the answers it can, translate the questions into plain language, and ask you for what’s missing. Use it to complete one form or build a helper others can use.

For the user

Share the form with your AI as a webpage link, PDF, document, image, or whatever format you have, and paste the prompt below. It will ask whether you want to fill it out once or create a tool for other people.

For your own form, you’ll supply answers the AI cannot find and confirm or correct every answer it finds, infers, or drafts, one at a time or in small related groups, before it treats them as final. You’ll also review the completed result. For a reusable helper, you’ll confirm who will use it and review the conversational experience the AI builds; each person using it will likewise confirm or correct every AI-supplied answer along the way before it becomes final.

Guidance for the AI

These are general instructions, not a description of this user's particular AI system, tools, access, or circumstances. Adapt the method to the user's goal and the capabilities you can actually use, while preserving the activity's essential sequence, review points, and intended result.

Start with the activity and the user's context

Read the complete activity and guide the user through it. Use the activity prompt yourself; the user does not need to paste it again. Keep the responsibilities of the user, the AI, and any colleague clear.

Begin with what the user has already supplied. Establish the intended result and resolve missing context through focused questions when needed. Work within the activity's stated scope, permissions, required confirmations, and stopping points, including any deliberate order for introducing information. Make progress with the available information and reasonable, clearly identified assumptions where the activity allows them.

Check the capabilities needed for this work

Inspect relevant available tools, connectors, and data. Use small, relevant, non-destructive checks to establish whether necessary access or a capability actually works. Naming a product does not establish access to it. Let this activity determine what needs checking; do not impose a universal preliminary questionnaire or a full system audit.

When useful access is missing:

• Explain what you cannot reach and how that material would help.
• Investigate practical setup, connection, export, or other access alternatives before asking the user to gather material or troubleshoot.
• If progress depends on access, request only the specific user action or permission needed to enable it or provide the material another way. Continue useful independent work while awaiting the response.
• Offer to walk the user through an export or upload. Welcome files, links, copied text, rough notes, or an interview as supported by your system and the activity, and use the best available route. If the missing access is optional, continue with what is available.

Ask focused questions and respect the answers

Ask questions only when necessary to complete the work well or satisfy the activity's required review points. Build on answers already supplied. Ask one question at a time, or a small group of closely related questions. Wait on work that depends on the answer, and continue useful work that does not. Explain unfamiliar language with examples where helpful.

Confirm answers or decisions you supply on the user's behalf before treating them as the user's final answers or decisions. This does not require separate approval of every sourced finding; follow the activity's specific review requirements.

Keep evidence and independent contributions distinct

Distinguish evidence, inference, and missing information. Check your work against the source material.

Use independent agents when your system actually supports delegation and independent investigation or review would improve the work:

• Give each agent a bounded brief and the context appropriate to its role.
• Keep discovery independent until synthesis where the activity requires it. Preserve any requirement for the user or colleagues to contribute before seeing other explanations.
• Follow the activity's stated fallback and any required confirmation when independent agents or fresh contexts are unavailable. Preserve reviewer isolation where required; describe shared-context or sequential passes accurately.

Produce, check, and revise the result

Do the substantive investigation and production. Carry the activity through to the most complete, usable result within its scope, rather than ending with a plan or handing the main work back to the user. When missing information does not prevent useful work, proceed with clearly identified assumptions where the activity permits them.

Check the result against the activity's completion criteria. At each substantive delivery, state what is complete and any material unfinished work, assumptions, helpful missing information, and remaining uncertainty. Explain how these affect what the user can use, and help the user revise the result.

Preserve the activity's stopping point. Do not turn a rehearsal, investigation, or test-design activity into execution that the user has not authorized.

Your AI prompt

Help me turn this form into a clear, easy conversation. Do as much of the work as you can yourself and keep my part manageable.

1. Get the form and establish the goal.

If I haven’t provided the form, ask for it. Accept a webpage link, PDF, document, image, or whatever format it comes in. Try to open and read it with your available tools. If you cannot access it, explain a workable way for me to share it, offering to walk me through downloading, exporting, or copying it.

Ask whether I want help filling out this form once or a reusable tool that helps other people fill it out. Wait for my answer. Read the whole form carefully, including:

• Instructions.
• Required fields.
• Answer formats.
• Questions that appear only after particular answers.

Identify what each question actually requires and how answers map back to the original fields. If part of the form is missing or unreadable, ask for that part before relying on it. Then follow the appropriate route below.

2. If I’m filling it out once, find the answers you can.

Use what you already know about me and inspect your available access to email, local files, and other connected documents. Search relevant sources for answers before asking me to supply them. If the form mentions a project, look for that project in accessible desktop files, email, and connected documents. Read enough context to establish that you have the right person, project, and current information.

Confirm every AI-supplied answer

Keep track of which source supports each answer. Use clearly supported facts to prepare draft answers; keep uncertain, conflicting, or outdated information for clarification. Whenever you find, infer, or draft an answer yourself, show it to me and ask me to confirm or correct it before treating it as final. Check these answers one at a time or in small related groups so the review stays manageable. Leave choices and declarations that require my judgment for me to answer. If useful access is missing, investigate available connection options, then ask for the specific action needed or continue with an interview.

Interview about what is missing

Organize the unanswered questions into a simple sequence. Ask questions that determine which later sections apply first, then group related topics. Ask one question at a time, or a small group that is naturally answered together, such as name, birthdate, and address. Wait for my response, record it, and use it to avoid asking again for information already supplied.

Preserve the meaning of the form

Translate confusing questions into plain language while preserving what the form is asking. Explain jargon and give a short example when it will help me answer. If a term has an official definition in the form or its guidance, use that definition. When the meaning remains unclear, identify the ambiguity rather than silently changing the question. Turn my conversational answers into the form’s required wording or format, checking any interpretation that could change my meaning.

Complete and review the form

Once you have the answers, put them into the original form if your tools allow. Otherwise produce a clearly labeled, field-by-field answer sheet I can transfer easily. Check:

• Required fields and conditional sections.
• Dates and answer lengths.
• Consistency.

Show me the completed draft, flag anything still unresolved, and ask for corrections. Apply them. Leave submission, signatures, and final declarations for my explicit decision.

3. If I want a reusable helper, choose a workable format and build it.

Establish who will use the helper and what they need at the end: for example, a completed PDF or answers ready to paste into an online form. Inspect your actual capabilities for creating interactive artifacts or web apps and making them available to those users. Explain the simplest workable approach and how people would open it and get their completed answers. Confirm that approach with me before building.

Build the conversation and confirmation rules

Build the helper around the form’s fields and branching rules. Use the same plain-language explanations, small questions, answer tracking, and final review described above. Any answer the helper finds, infers, or drafts must be shown to that person for confirmation or correction before it is treated as final. Let people correct earlier answers and update dependent questions accordingly. Use each person’s own answers or sources they choose to provide; give new users a fresh session. Keep a clear mapping from the conversation to the original form.

App or interactive artifact

If we choose an app or interactive artifact, create it and check representative paths through it, including an answer that reveals or skips a section, an incomplete answer, and a correction to an earlier answer where those cases apply. Verify that the resulting completed form or answer sheet matches what the person supplied. Fix problems and give me the working artifact, with a usable link or opening instructions. If hosting needs a decision or action from me, prepare the artifact first and explain the specific remaining step.

Reusable prompt paired with the form

If a reusable prompt accompanying the form would be the simplest useful approach with your available capabilities, choose that format and write the complete prompt. It must tell a new AI how to read that form, explain its questions, interview the person, track answers, and produce the completed output. Include the form-specific field mapping and branching instructions, and explain how to supply the original form. Check it with sample answers and revise gaps you find. Deliver the prompt and form together, or clear instructions for pairing them, so another person can begin without this conversation.

Done when

You have a completed form or an organized answer sheet ready for your final review, or a reusable conversational helper that lets other people produce their own answers.

These instructions were written in September 2026 using the frontier models available at the time. You may now have better tools or models, or restrictions that make some tasks harder. Your AI should adapt to the tools and access you actually have and find the best available way to help. Everything here is a starting suggestion for you and your AI, not a proven best practice for every setup. Feel free to rewrite any of it to better meet your needs.