With the rapid development of AI technology, from ChatGPT to the explosion of DeepSeek, AI can now genuinely play a role in assisting our daily work.
Beyond specific areas such as assisted programming and copywriting, we can also make full use of AI’s powerful logical capabilities and vast foundational data to help us propose practical and feasible improvement plans for some of the difficulties we encounter at work.
But for AI to give us actionable solutions, the design of the prompt framework and the contextual information provided to AI are extremely critical. A well-designed prompt framework can help us quickly obtain highly valuable references.
A Case of the Predicament We Face
Here we use a fairly common case of an agile team’s quality predicament to illustrate:
As a test engineer responsible for quality assurance in an agile team running the Scrum framework, the team also includes an SM, a PO, two Android developers (among whom Xiao Li is responsible for the overall system design of the App), one back-end developer, and one part-time visual designer. The team’s product is a companion Android App for an IoT smart bracelet product. However, due to issues such as coordination with the hardware team, frequently changing requirements, untimely visual design, delayed test submissions, and a backlog of bugs, product quality is not optimistic. The App has now gone through multiple iterations, and the planned launch date is drawing ever closer. How can we propose an improvement plan for the current quality predicament?
Designing the Prompt Framework
For AI to help us produce a feasible improvement plan, we first need to design the prompt framework reasonably, so that the final plan better fits the team’s current situation and is actionable.
The key points of the prompt framework design are then:
- A given role, so that AI determines from what perspective to propose the improvement plan
- The current status quo, that is, sufficient background information and context
- What AI should analyze in detail, or what its task should be
- The specific output expected from AI, including format requirements and the elements to be included
- Some relevant constraints, which have an important impact on whether the plan is feasible
Now, for the predicament above, I want AI to help me provide an overall quality improvement plan for the team that can be discussed at the retrospective meeting. Based on the above thinking, by inputting the prompt framework into DeepSeek, we get the following output:
Quality Improvement Plan (For the Iteration Retrospective)
1. Overall Quality Goals
“Within 2 iterations, raise automated test coverage to 40%, shorten defect-fix response time by 50%, compress regression testing time by 30%, and establish a quality defense line linking requirement changes with testing”
2. Specific Improvement Initiatives
| Initiative | Owner | Execution Plan | Evaluation Method |
|---|---|---|---|
| Requirement Freeze Mechanism - Lock core functional requirements 3 days before each iteration - Hardware integration requirements are jointly reviewed by the PO and the hardware team to form interface documents | PO, hardware team lead | Sign the requirements baseline document at the iteration planning meeting Share hardware interface documents via Confluence | Reduce requirement changes by 50% Interface document version change records are traceable |
| Automated Test Case Tiering Strategy - L0 (core business flows): mandatory full automation - L1 (high-frequency usage scenarios): prioritized for automation - L2 (edge scenarios): manual verification | Test Engineer (Lead) Android Developer (Support) | Use Allure to label case tiers Daily builds prioritize running L0 cases | 100% pass rate of L0 cases as the test-submission exit criterion Reduce automation execution time by 40% |
| Developer Self-Testing Responsibility - Code submitted by developers must include unit tests (coverage ≥60%) - API tests are maintained by back-end engineers in Swagger | Android Developer Back-end Engineer | Include unit tests as a code review checklist item Link Postman API cases with the code repository | Raise the smoke-test pass rate of submitted builds to 80% Reduce the API defect rate by 30% |
| Hardware Compatibility Mock Service - Develop a simulator based on the hardware protocol - Decouple automated testing from real devices | Test Engineer (Lead) Hardware team provides protocol documents | Develop a Bluetooth protocol simulator using Python Automatically start the Mock service during Jenkins builds | Reduce test blockages caused by hardware dependencies to zero Raise automation stability to 95% |
| Defect Severity Management - P0 defects: respond within 2 hours - P1 defects: must be fixed in the current iteration - Residual defects require the PO to sign off on the risk | Test Engineer (Classification) Developer (Fix) | Configure automated defect workflow rules in Jira Sync the progress of the top 3 defects at the daily stand-up | Shorten the average defect-fix cycle to 8 hours Reduce the number of backlogged defects by 60% |
3. Risks and Responses
| Risk | Response |
|---|---|
| Developers resist unit testing | - Xiao Li (the system designer) demonstrates writing unit tests - Incorporate test coverage into the DoD completion criteria |
| The hardware protocol changes frequently | - Require the hardware team to notify of protocol changes 3 days in advance - Design a version compatibility mechanism for the Mock service |
| UE design delays affect UI testing | - Agree with the UE designer: deliver UI element ID specifications together with the visual drafts - Use a screenshot comparison tool with a 10% pixel-difference tolerance |
4. External Support Requirements
| Type | Specific Requirements |
|---|---|
| Tools | - Apply for a BrowserStack mobile cloud testing platform license - Purchase an Xray test case management plugin license |
| People | - The hardware team designates a protocol liaison - A DevOps engineer optimizes the CI/CD pipeline |
| Training | - Invite a unit testing expert to run a TDD workshop - Internal training on advanced Appium usage |
5. Effect Evaluation Milestones
| Milestone | Verification Criteria |
|---|---|
| Mid-iteration 5 | - L0 automated cases cover all core business flows - Unit test coverage ≥40% |
| End of iteration 5 | - Automation coverage ≥25% - Regression testing time ≤8 hours per run |
| End of iteration 6 | - Automation coverage ≥40% - Zero fatal defects in the production environment |
Notes on Key Execution Points:
- Daily Quality Dashboard: Hang the automation execution rate / defect trend chart in the team war room
- Quality Gate Mechanism: Code merge requests must pass L0 cases plus SonarQube checks
- Shift-Left Testing Practice: Test engineers produce a test-point matrix as soon as they participate in requirement reviews
This plan requires the whole team to sign a quality commitment letter. It is recommended that, at the iteration retrospective, the team use the “5 Whys analysis” to reach a consensus on improvement regarding the current quality issues.
As we can see, we have obtained an improvement plan that is highly targeted and quite feasible. Based on the team’s current situation, we can actually put it into implementation after some fine-tuning.
So making good use of AI prompt frameworks can not only help us get assistance on specific tasks, but also provide tremendous support for overall improvement initiatives of this kind.
Reply with “prompt framework” to get the complete prompt example shown above. Regarding more practical applications of AI at work, everyone is also welcome to leave comments or join the group to discuss!

