Clarify Vague Work Requests with Team Collaboration AI: Requirements, Completion Criteria, and Clarifying Questions
Practice turning a short work request from a team lead into a draft request the team can agree on again, separating confirmed requirements from missing conditions instead of letting AI fill in the gaps arbitrarily. Using Team Haneul’s request, 'Please prepare some workshop materials,' we leave completion criteria, included and excluded scope, and submission conditions as clarifying questions.
Content checked 2026.09.22Copyable prompts
Show contents
Who this is forNon-developer team members and team leads who want to reduce inconsistent interpretations caused by short messenger instructions or verbal requests
What you need
An AI chat tool that accepts text input
Prepare the original work request separately from background information that has already been confirmed
Be able to ask the requester again instead of arbitrarily deciding unknown conditions
Use fictional data instead of real confidential or personal information
01Surface interpretation gaps before acting on a short request
If Minseo sends Doyun the message 'Please prepare some workshop materials,' it is unclear whether the materials are for presentation or distribution, how much detail to cover, what format to deliver, or what should count as complete and by when. If the assignee starts by guessing, their completion criteria may diverge from the requester’s.
Here, the role of team collaboration AI is not to imagine what belongs in the blanks, but to separate confirmed information from missing requirements and create a list of questions. This article focuses on having the requester and assignee agree on the same completion criteria rather than on writing an individual prompt.
Item
Can it be known now?
How to handle it
Work purpose
Partly
Keep only the background that it is event material and confirm the detailed purpose
Deliverable type
Unknown
Ask whether it is presentation material, handout material, or something else
Completion criteria
Unknown
Ask about review criteria and who approves
Included scope
Unknown
Ask about required content and exclusions
Submission format and location
Unknown
Ask about file format and delivery location
Submission deadline
Unknown
Ask for the actual submission deadline separately from the event and instructor schedule
02Copy in only the original request and already confirmed background
Give AI only the original message, facts that have already been confirmed, and assigned roles. Avoid adding interpretation so the result can clearly distinguish confirmed information from items that should remain questions.
Sample input
Fictional team: Team Haneul
[Team-Role-1] Minseo: team lead · overall schedule
[Team-Role-2] Junho: venue · supplies
[Team-Role-3] Seoyeon: promotion · registration
[Team-Role-4] Doyun: program · instructor outreach
[Event-1] Event: Neighborhood Digital Basics Workshop
[Event-2] Event date: 2026-11-14
[Event-3] Event time: 14:00–16:00
[Event-4] Capacity: 30 people
[Event-5] Audience: local residents
[Event-6] Fee: none
[Venue-1] Option A: Civic Center 3rd-floor seminar room, capacity 32 people, projector available, elevator available, booking availability confirmed
[Venue-2] Option B: district library auditorium, capacity 60 people, booking availability not confirmed
[Schedule-1] Venue decision deadline: 2026-10-24
[Schedule-2] Final promotional material deadline: 2026-10-28
[Schedule-3] Instructor confirmation deadline: 2026-10-30
[Registration-1] Registrations as of 2026-10-23: 22 people
[Request-Minseo-1] Original message Minseo sent to Doyun: "Please prepare some workshop materials."
[Status-Doyun-1] Doyun is responsible for the program and instructor outreach, but the original message does not state what type of material to prepare.
[Status-Doyun-2] The original message does not state the amount of material, file format, content to include, content to exclude, submission location, submission deadline, or review criteria.
[Status-Doyun-3] The instructor confirmation deadline of 2026-10-30 is confirmed, but there is no record that this is the material submission deadline.
The final sentence is especially important. A related date should not automatically be turned into the material submission deadline. Instructor confirmation and material submission are different events, so if the original request contains no basis linking them, the submission deadline should remain 'Needs confirmation'.
03Ask AI to structure gaps and questions instead of filling requirements in
Instead of asking AI for a completed work instruction, ask it to separate what can currently be confirmed from what still needs verification. Make the clarifying questions short and specific enough to copy into an actual conversation.
Prompt
Using only the input above, turn Minseo’s request "Please prepare some workshop materials" into a clear draft work request.
Do not arbitrarily decide anything AI does not know, and leave every blank item as 'Needs confirmation'.
Output order:
- Currently confirmed background and assignee
- Request purpose
- Requirements
- Completion criteria
- Included scope
- Excluded scope
- Submission format · location
- Submission deadline
- Questions to ask Minseo
- What Doyun may start before receiving answers and what he should not start
Rules:
1) Use only names, dates, times, and numbers that appear in the input.
2) Do not turn the instructor confirmation deadline into the material submission deadline.
3) Presentation material, handouts, file format, amount, and required content are not in the input, so do not confirm them.
4) Do not invent an assignee or approver that is not in the data.
5) Whenever possible, attach [Request-Minseo-1], [Status-Doyun-Number], [Event-Number], or [Schedule-Number] at the end of each item.
6) At the end, propose a short checklist for the requester and assignee to review together.
04Draft a work request that separates confirmed requirements from clarifying questions
The key point in the following edited example is that it deliberately leaves blanks unresolved. Even though it is formatted as a 'work request,' it does not fill in deliverables or deadlines that are not in the input. Doyun is confirmed as the assignee, but Minseo must be asked what must be produced for the work to count as complete.
Example result
[Example result · not an actual AI output]
Draft work request
Requester: Minseo [Team-Role-1] [Request-Minseo-1]
Assignee: Doyun [Team-Role-4] [Request-Minseo-1]
Confirmed background
- Event: Neighborhood Digital Basics Workshop [Event-1]
- Event date: 2026-11-14 [Event-2]
- Event time: 14:00–16:00 [Event-3]
- Capacity: 30 people [Event-4]
- Audience: local residents [Event-5]
- Instructor confirmation deadline: 2026-10-30 [Schedule-3]
Original request
- Minseo: "Please prepare some workshop materials." [Request-Minseo-1]
Request purpose
- It appears to be preparation of materials for use at the workshop, but the specific purpose of the materials Needs confirmation. [Request-Minseo-1] [Status-Doyun-1]
Requirements
- Material type: Needs confirmation. [Status-Doyun-1]
- Amount: Needs confirmation. [Status-Doyun-2]
- File format: Needs confirmation. [Status-Doyun-2]
Completion criteria
- Needs confirmation. The original request does not state review criteria. [Status-Doyun-2]
Included scope
- Needs confirmation. The original request does not state what content to include. [Status-Doyun-2]
Excluded scope
- Needs confirmation. The original request does not state what content to exclude. [Status-Doyun-2]
Submission format · location
- Needs confirmation. [Status-Doyun-2]
Submission deadline
- Needs confirmation. The instructor confirmation deadline of 2026-10-30 is confirmed, but there is no record that it is the material submission deadline. [Status-Doyun-3]
Questions to ask Minseo
- What will the materials be used for?
- What type of deliverable and file format are required?
- What content must be included, and what should be excluded?
- What state counts as complete?
- Who reviews and gives final confirmation?
- Where should the materials be delivered?
- What is the actual material submission deadline?
Before Doyun receives answers
- May start: organize already confirmed background information such as event date 2026-11-14, time 14:00–16:00, capacity 30 people, and audience local residents. [Event-2] [Event-3] [Event-4] [Event-5]
- Should not start: confirm the material type, amount, file format, included or excluded scope, or submission deadline without clarification. [Status-Doyun-2] [Status-Doyun-3]
This draft prepares the team to start a conversation rather than pretending the work request is complete. Once Minseo answers about the deliverable and completion criteria, those answers can be incorporated back into the request. Arbitrarily defining the deliverable before receiving an answer can lock in guessed requirements.
05Do not expand related schedules or roles into detailed instructions without evidence
A common risk when giving AI an ambiguous request is that naturally filled-in blanks can look like confirmed work instructions. The following example invents a deliverable format, amount, submission deadline, and reviewer that were not in the input.
Flawed result
[Common flawed result · not an actual AI output]
Doyun will create 20 presentation slides and prepare 2 types of participant handouts. He will send the draft to Junho for review by 2026-10-27 18:00 and submit the final version on 2026-10-30.
The input does not specify presentation slides or handouts as deliverable types, nor does it give an amount or draft submission time. There is also no record that Junho is the reviewer, and there is no basis for using the instructor confirmation deadline as the material final-submission deadline. The more specific such a result sounds, the more misleading it can be. To correct it, return every invented condition to 'Needs confirmation' and move it into the questions for Minseo.
Checklist
Human review checklist
- Has the wording of the original request been preserved?
- Are confirmed event facts separated from unresolved material requirements?
- Has the instructor confirmation deadline not been mistaken for the material submission deadline?
- Has AI avoided inventing the deliverable type, amount, file format, included scope, or excluded scope?
- Has no reviewer or approver absent from the data been added?
- Is it clear which items must not be confirmed before the clarifying questions are answered?
- When applying this to real work, has any data that violates the organization’s information security policy been kept out of the input?
06Confirm a shared standard after the requester and assignee fill in the answers
After sending the AI-generated question list, update the request with Minseo’s answers and confirm that Doyun understands them the same way. In particular, agree on what 'complete' means in one sentence and have AI organize only what has been agreed.
Minseo answers the deliverable type, purpose, included and excluded scope, and actual submission deadline.
Doyun reads back the completion criteria as he understands them and confirms them.
Keep unclear items as 'Needs confirmation' instead of forcing an answer.
Share the confirmed request in a designated location accessible to the team, while excluding real confidential and personal information from AI input.
If anything changes later, record what changed so it is not confused with the previous request.
Even a clarified request can become inconsistent again if team members use different wording. The next article, “Create a Shared Team AI Prompt: Version-Control the Rules for Organizing Meeting Notes”, explains how to create a shared prompt managed by the team rather than a personal prompt and how to keep a change history.
What to check yourself
Fictional input data and edited examples · no actual AI execution · checks for matching names, dates, times, numbers, and unresolved items
Check that every code block role uses an allowed value
Check that every article contains input, prompt, output, and bad_output
Use regular expressions to verify that names, dates, times, and numbers in the output exist in the input of the same article
Check that the instructor confirmation deadline is not asserted to be the material submission deadline
Check that the deliverable type, amount, file format, included scope, excluded scope, and submission location are not confirmed without evidence
Check that no sentence expresses effects as percentages or multiples
Verification limits
These are edited learning examples, not actual AI execution or actual team work results. For a real request, check the organization’s information security policy, and have the requester and assignee directly agree on requirements, completion criteria, and deadlines.
When introducing team collaboration AI for the first time, it is easier to manage if you compare tasks where multiple people repeatedly collect and verify information rather than starting with eye-catching features. Using 6 fictional tasks from Team Haneul, this guide describes expected benefits qualitatively, checks review burden and sensitive-information risk, and then narrows down the first use case. AI organizes the candidates, but the team makes the final choice.
When creating AI meeting notes, instead of polishing only one person's notes, practice combining records from multiple attendees and organizing decisions, action items, TBD items, and differences between records together with source tags. Using fictional Team Haneul meeting notes from 2026-10-20, this guide shows how to create a draft for team sharing without forcing different wording into a single version.
Practice using AI to collect weekly updates in different formats from each team member and reorganize them into completed work, work in progress, delays, dependencies, help needed, and items needing confirmation. Using Team Haneul reports from the same period, this guide creates an AI weekly report example that preserves both reference times and sources when numbers differ instead of deleting one value.
This guide explains how to reorganize scattered notes into background, current status, material locations, next actions, cautions, and TBD items so someone else can continue the work immediately when the owner is absent. Using a fictional case in which Seoyeon's promotion and registration work is handed over to Junho, it creates both an AI work handover document and questions the recipient should verify.
Practice comparing confirmed facts, evidence, opposing views, and unverified assumptions in one framework without letting AI make the conclusion when team members disagree. Using Team Haneul’s Venue Option A and Option B for the workshop, we identify the decision owner and deadline clearly and leave items for the team to verify before making the final judgment.
Learn how to turn team collaboration AI into a team asset that is reviewed and revised together rather than a set of individual prompts. We follow Team Haneul as it improves an AI meeting-notes prompt from v1.0 to v1.1 and checks it again with the same test input.
Learn how to check facts, omissions, sensitive information, owners, and approval status by role before sharing an AI weekly-report draft with the team. Separate the author, reviewer, and approver and record what evidence was checked so AI output is not mistaken for the team’s official conclusion.