plan-studyPlanner-guided

Prompt file imported from cirovladimir/study-planner (.github/prompts/plan-studyPlanner-guided.prompt.md). Copyright stays with the author.

Plan: Learning-Oriented Study Planner Prompt

Revise the existing VS Code Copilot prompt so it guides both the construction of a study planner and the learner’s development as an AI-assisted software engineer. The existing product scope is strong; the revision will add a recommended stack, explicit teaching checkpoints, concrete engineering habits, and learning-focused acceptance criteria.

Steps

1. Keep the prompt as a self-contained .prompt.md for VS Code Copilot and confirm that the project starts from a blank workspace.
2. Define the product goal: a practical local-first study planner for learning AI-assisted coding and software engineering.
3. Recommend React, TypeScript, and Vite, with a brief explanation focused on learnability, fast feedback, component design, type safety, and common frontend tooling.
4. Preserve the existing MVP workflows:
   • Create a learning goal
   • Build and edit a roadmap
   • Schedule and complete study sessions
   • Practice coding tasks
   • Record reflections and blockers
   • Review progress and determine the next activity
5. Define a simple domain model, seeded demo data, and local persistence behind a small service boundary.
6. Add a teaching contract for every implementation phase:
   • Explain the relevant concept before coding
   • Propose one small change
   • Ask the learner to inspect or predict behavior where useful
   • Implement the change
   • Run focused validation
   • Summarize what was learned and what to try next
7. Organize implementation into independently verifiable phases:
   • Project scaffold
   • Application shell
   • Data model and persistence
   • Dashboard and roadmap
   • Study sessions and coding practice
   • Reflection and progress
   • Mock AI assistant
   • Quality hardening
8. Require engineering practices such as separation of domain logic and UI, form validation, accessible interactions, empty/loading/error states, focused tests, and formatting, linting, typechecking, and build validation.
9. Keep the AI layer provider-agnostic and deterministic for the MVP. It must not require credentials or imply that a real AI provider is connected.
10. Add learning-oriented acceptance criteria, including the ability to explain the architecture, modify an existing feature, and understand the validation output.
11. Keep authentication, cloud sync, real AI integrations, repository integrations, code execution, notifications, sharing, and advanced analytics outside the MVP.

Relevant files

• .github/prompts/plan-studyPlanner.prompt.md — revise the current prompt after inspecting its latest contents.
• /memories/learning_habits.md — preserve the preference for planning first, clean workspace practices, and iterative MVP development.

Verification

1. Confirm the prompt remains valid Markdown and is copy-pasteable as a single VS Code Copilot prompt.
2. Confirm it recommends React, TypeScript, and Vite with a clear rationale.
3. Confirm each implementation phase includes a learning checkpoint and focused validation.
4. Confirm it contains concrete screens, interactions, data entities, seeded data, persistence behavior, mock AI boundaries, and acceptance criteria.
5. Confirm it clearly separates MVP requirements from future enhancements.
6. Confirm it does not require an API key, backend, or external AI provider.

Decisions

• Target VS Code Copilot.
• Keep the prompt in .github/prompts/.
• Recommend React + TypeScript + Vite.
• Use local-first persistence and a deterministic mock AI service.
• Make learning and explanation part of the implementation contract.
• Revise the prompt only; do not scaffold the application during this planning step.

The plan is saved in session memory at /memories/session/plan.md.