AITraining2U

Programs

Resources

Case Studies

Quick Links

Enquire Now
AI Prompt Library

30 AI Prompts for Project Managers

Copy-ready prompts that assume you will attach the requirements, the notes or the project data — built around PM frameworks (WBS, RAID, RACI, pre-mortem) so planning and reporting take minutes.

30 copy-ready prompts Works with Claude, Copilot & ChatGPT
AI prompts for project managers — a planning board with sticky notes and a laptop, Malaysia

Each prompt gives the AI a delivery-manager persona, references what you will attach, and applies a project framework with a defined deliverable. These prompts are tool-agnostic and optimised to work across all major AI assistants — Claude, Microsoft Copilot and ChatGPT (and Google Gemini). Replace the [bracketed placeholders] with your own details, and always review AI output before you act on it.

Planning and scoping

1

Work Breakdown Structure Builder

Role

Delivery manager

Attach

Describe the project goal

Task
  • Break the goal into phases
  • Define milestones per phase
  • List work packages under each milestone
  • Specify the deliverable per package
Constraints
  • Mutually exclusive, collectively exhaustive
  • Flag packages with unclear scope
Output

A work breakdown structure with milestones.

See example AI output
Phase 1: Discovery (Milestone: Requirements Signed Off)
- WP1.1 Stakeholder interviews leads to Deliverable: interview notes
- WP1.2 Requirements doc leads to Deliverable: signed-off BRD
Phase 2: Build (Milestone: MVP Complete)
- WP2.1 Core module dev leads to Deliverable: working MVP
- WP2.2 QA cycle leads to Deliverable: test report
[Flag] WP2.1 scope for the reporting module is unclear - confirm with sponsor before estimating.
2

Project Charter Draft

Role

Project planner

Attach

Name the project

Task
  • State the objective
  • Define scope in and out of bounds
  • List key stakeholders
  • Set success criteria, assumptions and constraints
Constraints
  • One page
  • Unambiguous scope boundaries
Output

A one-page project charter.

See example AI output
Project: Warehouse Inventory App
Objective: cut stock-count errors by 50% within Q4.
In scope: barcode scanning, live stock dashboard.
Out of scope: supplier ordering module.
Stakeholders: Ops Director (sponsor), Warehouse Leads, IT.
Success criteria: error rate under 2%, live by 30 Nov.
Assumptions: handheld scanners already owned.
Constraints: RM80k budget, 10-week timeline.
3

Prioritised Backlog From Requirements

Role

Backlog builder

Attach

Paste the requirements list

Task
  • Convert each requirement into a user-story item
  • Write acceptance criteria per item
  • Prioritise using value versus effort
Constraints
  • Independent items
  • Testable items
Output

A prioritised backlog of user stories with acceptance criteria.

See example AI output
1. As a shopper, I want saved payment methods so checkout is faster.
   AC: card saved after first purchase; removable in settings. Value: High, Effort: Low.
2. As a shopper, I want order tracking so I know delivery status.
   AC: status updates within 1hr of carrier scan. Value: High, Effort: Medium.
3. As an admin, I want a returns dashboard so refunds are faster.
   Value: Medium, Effort: Medium.
4

Task Estimate And Critical Path

Role

Estimator

Attach

Paste the task list

Task
  • Estimate relative effort per task
  • Sequence the tasks
  • Flag dependencies and the critical path
Constraints
  • State the basis of estimate
  • Mark low-confidence items
Output

A sequenced task list with dependencies and the critical path.

See example AI output
1. Design API (3d, basis: similar past sprint) leads to blocks Task 3
2. Set up test env (2d) leads to parallel
3. Build endpoints (5d, low confidence - scope unclear) leads to depends on Task 1
4. Integration testing (3d) leads to depends on Task 3
Critical path: 1 leads to 3 leads to 4 (11 days). Task 2 has slack.
5

Project Timeline Outline

Role

Schedule planner

Attach

Name the project and duration

Task
  • Lay out the phases across the duration
  • Set key milestones with target dates
  • Note dependencies
  • Add buffer around critical milestones
Constraints
  • Realistic
  • Protect the critical milestones
Output

A phased timeline with milestone dates and buffer.

See example AI output
Project: CRM Migration (12 weeks)
Phase 1 Discovery (Wk1-2) - Milestone: Data map signed off (Wk2)
Phase 2 Migration build (Wk3-7) - Milestone: Test migration complete (Wk7), 3-day buffer
Phase 3 UAT (Wk8-10) - Milestone: Sign-off (Wk10)
Phase 4 Go-live (Wk11-12) - Milestone: Cutover (Wk12), depends on UAT sign-off.

Status and reporting

6

Weekly Status Report Builder

Role

PMO reporter

Attach

Paste this week's project updates

Task
  • Summarise progress against plan
  • State RAG status with the reason
  • List risks, issues and decisions needed
  • Outline next steps
Constraints
  • Exceptions-based
  • Honest RAG rating
Output

A weekly status report.

See example AI output
Project: Mobile App Revamp - Week 14
Status: AMBER (reason: design sign-off 4 days late)
Progress: 60% of sprint 6 complete, on track for feature freeze.
Risks: app store review may slip launch by 1 week.
Decisions needed: approve revised launch date of 15 Aug.
Next steps: finalise design by Fri, start regression testing Mon.
7

RAG Status One-Liner

Role

Status summariser

Attach

Paste project notes and name the project

Task
  • Assign a RAG colour
  • Justify the colour in one line
  • State the single most important fact for the reader
Constraints
  • Defensible colour
  • No green-washing
Output

A one-line RAG summary with justification.

See example AI output
Project: Payroll System Upgrade - AMBER
Justification: two of five test cycles failed due to tax-table errors.
Most important: go-live on 1 Sept is at risk unless fixes are verified by 20 Aug.
8

Executive One-Liner And Summary

Role

Executive communicator

Attach

Name the project

Task
  • Write a one-line status headline
  • State current position
  • State the key risk
  • State the ask
Constraints
  • Lead with the headline
Output

A one-liner plus a three-sentence summary.

See example AI output
Headline: on track, but vendor delay threatens the Sept launch.
Summary: development is 85% complete and testing starts next week. The hardware vendor has slipped delivery by 10 days, which could push go-live past the board deadline. We need approval to expedite-ship at an extra RM12k to protect the date.
9

Audience-First Stakeholder Update

Role

Stakeholder-update writer

Attach

Attach the project data

Task
  • Lead with what matters most to this audience
  • Summarise progress
  • Flag risks
  • State the decision needed
Constraints
  • Audience-first
  • Concise
Output

A stakeholder update.

See example AI output
For the Finance Committee - Budget System Rollout
What matters to you: spend is currently 4% under budget.
Progress: 3 of 4 modules live, final module in UAT.
Risks: currency-conversion bug found, fix in progress.
Decision needed: approve 1-week extension to finish regression testing before sign-off.
10

Delay Explanation Message

Role

Delay communicator

Attach

Name the project and the delay details

Task
  • State what slipped and why
  • Explain the impact
  • Present the recovery plan
  • Give the revised date
Constraints
  • Candid
  • No blame
  • Credible plan
Output

A delay-explanation message.

See example AI output
Subject: update on Website Relaunch timeline
The content migration ran two weeks behind schedule due to a larger-than-expected volume of legacy pages needing rework. This pushes the relaunch from 1 Aug to 15 Aug.
Recovery plan: adding one contractor and running migration and QA in parallel from next week.
Revised go-live: 15 August. We will confirm progress at Friday's checkpoint.

Risk and issues

11

RAID Log Builder

Role

Risk manager

Attach

Describe the project context

Task
  • List risks with likelihood, impact and mitigation
  • List assumptions
  • List issues with owners
  • List dependencies
Constraints
  • Specific
  • Prioritised
  • Owned
Output

A RAID log covering risks, assumptions, issues and dependencies.

See example AI output
Risks: vendor API instability (Likelihood: Med, Impact: High, Mitigation: build fallback queue) - Owner: Tech Lead
Assumptions: client will provide test data by Wk3
Issues: staging environment down since Mon - Owner: DevOps, due Wed
Dependencies: legal review must complete before the vendor contract is signed.
12

Single Risk-Register Entry

Role

Risk-entry drafter

Attach

Name the specific risk

Task
  • Describe the risk and its trigger
  • Rate likelihood and impact
  • List current controls
  • State response strategy and owner
Constraints
  • Measurable where possible
Output

A single risk-register entry.

See example AI output
Risk: key developer resignation mid-sprint
Trigger: notice period started
Likelihood: Low (15%) | Impact: High (2-week slip)
Current controls: pair programming on critical modules
Response: mitigate - cross-train a second engineer within 2 weeks
Owner: Engineering Manager
13

Issue Triage Action Log

Role

Issue triager

Attach

Paste the list of open issues

Task
  • Prioritise issues by impact and urgency
  • Unblock the critical path first
  • Assign owners and due dates
Constraints
  • Prioritise by impact and urgency
  • Unblock critical path first
Output

A prioritised action log.

See example AI output
1. Payment gateway timeout (blocks launch) - Owner: Backend Lead, Due: Thu
2. Broken analytics tracking - Owner: Data team, Due: Fri
3. Typo in confirmation email - Owner: Content, Due: next sprint
Critical path unblocked once Item 1 is resolved.
14

Project Pre-Mortem

Role

Pre-mortem facilitator

Attach

Name the project

Task
  • Assume the project has failed
  • List likely failure reasons ranked by probability
  • Give a preventive action for each
Constraints
  • Brutally honest
  • Actionable
Output

Pre-mortem findings with preventive actions.

See example AI output
Project: New Branch Opening (assume failed)
1. (40%) Staff not hired in time leads to Prevention: start recruitment now, not at Wk8
2. (30%) Landlord fit-out delays leads to Prevention: lock a penalty clause into the lease
3. (20%) Local permits rejected leads to Prevention: submit the permit application this week, not next month
15

Issue Resolution Options

Role

Issue analyst

Attach

Paste the issue details

Task
  • Summarise the issue
  • Propose three resolution options
  • State cost, time and risk of each
  • Give a recommendation
Constraints
  • Quantify the trade-offs
Output

Resolution options with a recommendation.

See example AI output
Issue: database performance degraded 40% after last release.
Option A - Roll back: cost RM0, time 1hr, risk: loses new features.
Option B - Add indexing: cost RM5k, time 2 days, risk: low, does not fully fix.
Option C - Re-architect query layer: cost RM40k, time 3 weeks, risk: high but permanent fix.
Recommendation: Option A now to stabilise, then Option C planned for next quarter.

Stakeholder communication

16

Stakeholder Map And Comms Plan

Role

Stakeholder strategist

Attach

Name the project

Task
  • List stakeholder groups
  • Rate interest and influence for each
  • Note what each group cares about
  • Set communication approach and cadence
Constraints
  • Actionable
  • Prioritise high-influence high-interest
Output

A stakeholder map with a communications plan.

See example AI output
Project: ERP Rollout
Finance Director - Influence: High, Interest: High. Cares about cost control. Comms: weekly 1:1.
End users (Ops team) - Influence: Low, Interest: High. Cares about ease of use. Comms: bi-weekly newsletter.
IT Security - Influence: High, Interest: Med. Cares about compliance. Comms: monthly steering update.
17

Project Kickoff Email

Role

Kickoff facilitator

Attach

Name the project

Task
  • State the objective and scope
  • List roles and the timeline
  • Set ways of working
  • Give the first action
Constraints
  • Energising
  • Clear
Output

A kickoff email.

See example AI output
Subject: kicking off Project Nova - let's build!
Objective: launch the loyalty app by 1 Oct.
Scope: iOS/Android app plus points engine; excludes in-store kiosks.
Roles: Aisyah (PM), Ravi (Tech Lead), Mei (Design).
Timeline: 8 weeks, sprint reviews every Friday.
Ways of working: async updates in Slack, stand-up 9:15am daily.
First action: everyone reviews the requirements doc by Wed.
18

Expectation-Reset Message

Role

Expectations manager

Attach

Describe the scope or timeline change

Task
  • State the change
  • Explain the reason
  • Describe the impact
  • Present options and a recommendation
Constraints
  • Proactive
  • Honest
  • Solution-oriented
Output

An expectation-reset message.

See example AI output
Subject: change to the Analytics Dashboard scope
The change: real-time data feed is being descoped from Phase 1.
Reason: source system API is not ready until Q1 next year.
Impact: dashboard will show data with a 24-hour lag at launch.
Options: (1) launch with daily refresh now, add real-time later, or (2) delay launch 3 months.
Recommendation: Option 1 - ship value now, upgrade later.
19

Technical-To-Business Translator

Role

Translator

Attach

Paste the technical update

Task
  • Read the technical update
  • Extract the business impact
  • Extract the decision needed
  • Rewrite in plain language
Constraints
  • Accurate
  • No jargon
Output

A business-friendly update.

See example AI output
Technical note: API rate-limiting caused 502 errors during peak load, mitigated via a caching layer.
Business version: some customers saw the app fail to load during our busiest hour last week. We fixed it by storing frequently used data closer to users. No customer data was affected, and we recommend no further action at this time.
20

Steering Committee Talking Points

Role

Governance-prep assistant

Attach

Name the project

Task
  • Summarise current status
  • List decisions needed
  • List risks to escalate
  • Give a recommendation for each
Constraints
  • Decision-focused
  • Concise
Output

Steering-committee talking points.

See example AI output
Project: Data Centre Migration
Status: 70% complete, on budget.
Decision needed: approve additional RM50k for backup hardware - Recommend: approve, risk of data loss otherwise.
Risk to escalate: sole networking vendor has a 3-week lead time - Recommend: approve dual-sourcing now.
Next milestone: cutover rehearsal, 2 Aug.

Meetings and coordination

21

Meeting Agenda Designer

Role

Meeting designer

Attach

Name the meeting type and project

Task
  • State the meeting objective
  • List required pre-read
  • Time-box each agenda item
  • Define the outcome to leave with
Constraints
  • Focused
  • Realistic timing
Output

A time-boxed meeting agenda.

See example AI output
Sprint Planning - Project Falcon (60 min)
Objective: commit to Sprint 9 scope.
Pre-read: backlog groomed items, capacity sheet.
0-10min: review last sprint's velocity
10-40min: walk backlog, size top items
40-55min: confirm sprint goal and commitments
55-60min: risks and blockers
Outcome: agreed sprint backlog with owners assigned.
22

Meeting Notes To Actions

Role

Minute-taker

Attach

Paste the raw meeting notes

Task
  • Extract decisions made
  • Extract actions with owners and due dates
  • Extract open questions
Constraints
  • Unambiguous ownership
  • Nothing lost
Output

Three sections: decisions, actions, open questions.

See example AI output
Decisions: proceed with Vendor B for hosting.
Actions:
- Send contract to Legal - Owner: Priya, Due: Wed
- Update budget sheet - Owner: Farhan, Due: Fri
Open questions: does Vendor B support our SSO requirement? Need an answer before the contract is signed.
23

Stand-up Summary Builder

Role

Stand-up summariser

Attach

Paste the team's individual updates

Task
  • Surface blockers first
  • Summarise progress
  • Identify who needs help
Constraints
  • Surface blockers first
Output

A stand-up summary.

See example AI output
Blockers: Ravi blocked on staging access (needs IT ticket approved).
Progress: login flow done and merged; checkout UI 80% complete; QA started regression on payments.
Needs help: Mei needs a design review slot today to unblock checkout.
24

Acceptance Criteria Writer

Role

Requirements analyst

Attach

Describe the feature or task

Task
  • Write acceptance criteria for the feature
  • Use given-when-then format where it fits
  • Ensure each criterion is verifiable
Constraints
  • Unambiguous
  • Verifiable
Output

A set of acceptance criteria.

See example AI output
Feature: password reset via email
Given a registered user requests reset, When they submit a valid email, Then a reset link is sent within 60 seconds.
Given the link is older than 30 minutes, When clicked, Then the user sees 'link expired' and can request a new one.
Given an invalid email is submitted, When the form is submitted, Then a generic message is shown with no account enumeration.
25

Overdue Action Chase Messages

Role

Follow-up coordinator

Attach

List the people and their overdue actions

Task
  • Draft one message per person
  • Reference the specific overdue action
  • Set a firm but friendly deadline
Constraints
  • Not heavy-handed
  • Clear ask
Output

One chase message per person.

See example AI output
To Aisyah: Hi Aisyah, checking in on the vendor comparison sheet due last Friday - could you share it by end of day tomorrow so we can finalise the choice at Thursday's review?
To Daniel: Hi Daniel, the test cases for the payments module were due Monday. Can we get them by Wed morning so QA is not blocked?

Retrospectives and improvement

26

Retrospective Session Designer

Role

Retro facilitator

Attach

Name the project or sprint

Task
  • Choose a retro format
  • Write questions for what went well, what did not and what to change
  • Design how findings become commitments
Constraints
  • Psychologically safe
  • Action-producing
Output

A retrospective plan with questions.

See example AI output
Format: Start-Stop-Continue, 45 min, anonymous input first.
Questions: what should we start doing? What should we stop? What should we continue?
Safety note: no names attached to raised issues, facilitator reads items aloud.
Commitments: top 3 voted items become actions with an owner and a check-in date at the next retro.
27

Retro Feedback Theme Synthesiser

Role

Feedback synthesiser

Attach

Paste the raw retro feedback

Task
  • Cluster feedback into themes
  • Quantify how often each theme appeared
  • Produce three concrete improvement actions with owners
Constraints
  • Theme over anecdote
  • Actionable
Output

Themes with quantities and three improvement actions.

See example AI output
Theme 1: unclear requirements (mentioned 6x)
Theme 2: too many meetings (mentioned 4x)
Theme 3: slow code review turnaround (mentioned 5x)
Actions:
1. Add a requirements checklist before sprint start - Owner: PM
2. Cut recurring meetings from 5 to 3 per week - Owner: Team Lead
3. Set a 24-hour SLA for code reviews - Owner: Eng Lead
28

Lessons-Learned Documentation

Role

Knowledge manager

Attach

Name the project

Task
  • Document what worked
  • Document what to avoid
  • Give specific recommendations for future teams
Constraints
  • Concrete
  • Reusable
Output

Lessons-learned documentation.

See example AI output
Project: Office Relocation
What worked: early vendor walk-throughs caught 3 issues before signing the lease.
What to avoid: do not finalise furniture orders before IT confirms the desk-power layout - caused rework.
Recommendations: build a 2-week buffer before any move date; assign one single point of contact for building management liaison from day one.
29

Delivery Slip Root-Cause Analysis

Role

Delivery analyst

Attach

Describe the slipped project or milestone and its context

Task
  • Analyse why the milestone slipped
  • Identify systemic causes, not just symptoms
  • Recommend process changes
Constraints
  • No blame
  • Focus on the system
Output

A root-cause analysis with recommended process changes.

See example AI output
Milestone: UAT sign-off slipped 3 weeks.
Symptom: testers found requirements ambiguous.
Systemic cause: no formal requirements review gate before build started.
Another cause: test environment provisioned only after build finished, compressing test time.
Process changes: add a requirements review gate before Sprint 1; provision test environments in parallel with build, not after.
30

Improvement Ideas Action Plan

Role

Improvement planner

Attach

Paste the list of improvement ideas

Task
  • Assess effort and expected benefit per idea
  • Assign owners
  • Sequence quick wins first
Constraints
  • Realistic
  • Sequence quick wins first
Output

A prioritised improvement action plan.

See example AI output
1. Standardise sprint templates (Effort: Low, Benefit: Med) - Owner: PM - Do first
2. Automate status report generation (Effort: Med, Benefit: High) - Owner: PMO - Next
3. Restructure the team into two pods (Effort: High, Benefit: High) - Owner: Head of Delivery - Plan for next quarter

Frequently Asked Questions

Each applies a project framework — WBS for scoping, RAID for risk, pre-mortem for foresight, RACI for accountability — and assumes you will attach the requirements, notes or project data. That turns hours of documentation and reporting into minutes, so you focus on the coordination and judgement that keep projects on track.

Yes — the prompts work from your attached requirements, status notes and project data. Use approved enterprise AI for anything confidential, and anonymise client or commercial details where you can.

Copilot is convenient inside Microsoft Project, Teams and Planner; Claude and ChatGPT are strong for planning, summarising and drafting. Many PMs also connect AI to their PM tools. These prompts work across all of them.

It drafts a solid first-pass plan, work breakdown and risk log from your inputs, which you refine with real-world judgement. Treat it as an accelerator, not autopilot.

Yes. AITraining2U's courses are HRD Corp SBL-KHAS claimable for eligible Malaysian employers.

Yes. Give the AI your task or work-package list plus the team roles, and ask it to assign Responsible, Accountable, Consulted and Informed for each item, then flag any task with no owner or more than one Accountable. This surfaces accountability gaps fast, a common cause of stalled decisions. It works well combined with a stakeholder map: use the RACI for day-to-day task ownership and the stakeholder map for communication cadence. Review the AI's first draft with your team, since only they know the true reporting lines and informal influence a matrix alone will not capture.

AI will not replace the conversation, but it removes the admin around it. Before planning, ask it to size and sequence a groomed backlog from your notes. During the sprint, feed it daily updates and it drafts the stand-up summary with blockers surfaced first. After the sprint, paste retro feedback and it clusters it into themes with suggested actions in seconds instead of the usual sticky-note exercise. The team still discusses and decides; AI just handles the synthesis and formatting so ceremonies spend more time on judgement and less on paperwork.

Increasingly, yes. Copilot is built into Microsoft Project, Planner and Teams. Claude and ChatGPT can connect to Jira, Asana, Trello and similar tools through their connector or plugin ecosystems, letting the AI read tickets and draft updates without you exporting data manually. If your organisation has not enabled a connector yet, the fallback is simple: export or copy the relevant board, backlog or issue list and paste it into the prompt. Either way, always check your company's data-handling policy before connecting AI to a tool that holds client or commercial information.

Treat AI the same way you would treat an external contractor with no NDA by default. Use your organisation's approved enterprise AI tool where chats are not used for training and data stays within agreed boundaries, rather than free consumer accounts. Strip or replace client names, contract values and personal data with placeholders before pasting anything in, and only attach the minimum data needed to get a useful answer. For genuinely sensitive material, such as unsigned contracts, personal data or security details, keep it out of any AI tool entirely, enterprise or not, unless your legal and IT teams have specifically cleared that use case.

Yes. Give it your team roster, their current allocations and the upcoming task list, and ask it to flag over-allocated people, identify who has spare capacity, and suggest a rebalanced plan. It is also useful for scenario-testing: ask what happens to the timeline if one person is pulled onto another project for two weeks. The AI is only as good as the data you feed it, so keep allocation numbers current, and always sanity-check its suggestions against real skills and availability, since a spreadsheet does not know who is on leave or who is the only person who understands the legacy system.

Go beyond prompts — train your team

Prompts are the start. AITraining2U runs hands-on, HRD Corp SBL-KHAS claimable AI training for Malaysian teams — from everyday AI productivity to building agents that run project management workflows end-to-end.