AI Unified Process - stack-agnostic core methodology (requirements, entity model, use cases)
92
95%
Does it follow best practices?
Impact
92%
1.29xAverage score across 12 eval scenarios
Low
Low-risk findings worth noting
Your team has just completed the requirements phase for a new user account system. The product owner has signed off on the requirements catalog at docs/requirements.md and a high-level use case diagram at docs/use_cases.puml that captures two core flows: account registration and login.
The engineering team is now ready to begin implementation, but they need detailed use case specifications before writing any code. The specifications will serve as the contract between the business and the development team — they must cover both the happy path and the various ways things can go wrong, so that everyone from stakeholders to developers can agree on exactly how each interaction should behave.
The team documents its systems with the AI Unified Process, so the specifications must conform exactly to the standard AI Unified Process use case specification format — file naming convention, section structure, business-level step language, and globally unique business rule IDs. Specifications that deviate from the standard format will be rejected during review.
Write complete use case specification documents for all use cases shown in docs/use_cases.puml. Place the files in the docs/use_cases/ directory, with one file per use case.
For each use case, the specification must cover: a complete overview section, preconditions, a fully numbered main success scenario, alternative flows for the meaningful error and edge cases, success and failure postconditions, and any relevant business rules.
.tessl-plugin
evals
scenario-1
scenario-2
scenario-3
scenario-4
scenario-5
scenario-6
scenario-7
scenario-8
scenario-9
scenario-10
scenario-11
scenario-12
skills