Input: Jira issue description and project goalsA document describing the game mechanics from contexts

AI Task: Extract requirements for the API:

• Extract key aspects: rules, user flows, data flows, core functionality, and functionalities.Highlight constraints etc.

• Create a glossary of relevant terms and definitions.

• Determine roles/entities involved (e.g., user, performance, securityadmin, NPC, etc.).

Output: A structured table of API requirements.concise, structured overview covering:

• Fundamental concepts and components

• Main interactions or workflows

• Basic relationships among these elements

Input: API requirements and data flow contextOutput from "Requirements Analysis" (glossary, core components, etc.)

AI Task: Define the endpoints for the API:

• Compile a full list of possible user actions (in a game context: movement, item usage, etc.; in software: CRUD actions, triggers, etc.).Specify request/response structures

• For each action:

  • Name (short tag)

  • Short description

  • Requirements/conditions (e.g.,

  headers

  • parameters,

  body

  • permissions,

  query params).

  • items)

  • Result (changes in system state, UI updates, etc.)

Output: A structured table of API endpoints.Comprehensive list (table or hierarchy) of all user actions with detailed descriptions

Input: API design and endpoint detailsUser actions and entities

AI Task: Create developer-friendly API documentation:

• Include endpoint descriptions, examples, and error handling.

• Suggest tools like Swagger/OpenAPI for documentation generation.

Output: A detailed API documentation draft.

Input: Full API design, endpoints, and documentation

AI Task: Review the API design for completeness:

• Identify gaps or inconsistencies in endpoint definitions.

• Suggest improvements for scalability or security.

Output: A validation report with actionable feedback.

• Identify main entities (e.g., Player, User, Item, Resource, etc.) and their fields (id, status, relationships, etc.).

• Determine client-side vs. server-side storage (or local vs. cloud for software).

• Define how and when the system synchronizes data.

Output: A schema (ER diagram or tabular format) clarifying how data is structured and updated

Input: User actions and the data model

AI Task:

• Map endpoints or service methods to each action (REST, GraphQL, WebSocket, etc.).

• Specify request/response formats (HTTP method, path, parameters, body, etc.).

Output: A table with next columns:

• Endpoint: API route.

• HTTP Method: GET, POST, PUT, DELETE.

• Request Body: JSON schema for the request payload (use `json` formatting).

• Response Structure: JSON schema for the response payload (use `json` formatting).

Input: Draft API specification and the list of user actions

AI Task:

• Verify that all actions are covered (no duplicates or gaps).

• Adjust data structures or endpoints if needed.

Output: A refined API spec addressing all requirements and potential concerns