#runSubagent

.md

Please implement this refactor plan: #file:[refactorplan.md]. Analyze the pending tasks & todos listed in the document and plan out how to split them up into subtasks.

For each task, spawn an agent using #runSubagent, and ensure you orchestrate them properly. It is probably necessary to run them sequentually to avoid conflicts, but if you are able, you are encouraged to use parallel agents to speed up development. For example, if you need to do research before starting the implementation phase, consider using multiple parallel agents: one to analyze the codebase, one to find best practices, one to read the docs, etcetera.

You have explicit instructions to continue development until the entire plan is finished. do not stop orchestrating subagents until all planned tasks are fully implemented, tested, and verified up and running.

Each agent should be roughly prompted like so, adjusted to the selected task:

[TASK DESCRIPTION/INSTRUCTIONS HERE]. Ensure you read the refactor plan & agents.md; keep both files updated as you progress in your tasks. Always scan the repo & documentation for the current implementation status, known issues, and todos before proceeding. DO NOT modify or create `.env`: it's hidden from your view but has been set up for development. If you need to modify env vars, do so directly through the terminal. 

Remember to use `uv` for python, eg `uv run pytest`, `uvx ruff check [path]`, etc.  Before finishing your turn, always run linter, formatter, and type checkers with: `uvx ruff check [path] --fix --unsafe-fixes`, `uvx ty check [path]`, and finally `uvx ruff format [path]`. If you modified the frontend, ensure it builds by running `pnpm build` in the correct directory. 

Once done, atomically commit the changes you made and update the refactor plan with your progress.

Test improved version:

You are working inside an existing Django codebase.

Your task is to fully implement the requested work by following a structured, plan-first workflow and maintaining production-quality standards.

PHASE 1 — Project Awareness

• Always read .github/copilot-instructions.md first.
• Scan the repository structure, README, docs, and existing code to understand:
  • Architecture and patterns
  • Current implementation status
  • Known issues and TODOs
• If .github/copilot-instructions.md is outdated, incomplete, or incorrect:
  • Update it to reflect reality
  • Remove irrelevant or misleading guidance

PHASE 2 — Task Planning

• Create or update task_plan.md.
• Translate the provided issues/problems into:
  • Clear tasks
  • Smaller, actionable subtasks
• Explicitly note:
  • Dependencies between tasks
  • What can be done sequentially vs independently
• Do NOT start coding until the plan is complete and coherent.

PHASE 3 — Execution (Simulated Roles)
For each task, spawn an agent using #runSubagent, and ensure you orchestrate them properly. Proceed step-by-step using this mental model:

• Analyzer: understand constraints, edge cases, and existing code impact
• Researcher (if needed): confirm best practices or library usage
• Implementer: write clean, idiomatic code
• Reviewer: refactor for clarity, correctness, and maintainability
• Custom: Any additional agent that might be necessary to complete task.

Each agent should be roughly prompted like so, adjusted to the selected task:

[TASK DESCRIPTION/INSTRUCTIONS HERE]. Ensure you read the refactor plan & agents.md; keep both files updated as you progress in your tasks. Always read .github/copilot-instructions.md file then scan the repo & documentation for the current implementation status, known issues, and to-dos before proceeding. DO NOT modify or create `.env`: it's hidden from your view but has been set up for development. If you need to modify env vars, do so directly through the terminal.
Before finishing your tasks, always check linter, formatter, typing so everything would follow best practices.

Update task_plan.md as tasks are completed.

PHASE 4 — Quality Gates (Mandatory)
Before considering a task complete:

• Code must follow best practices
• Run and fix:
  • Linter issues
  • Formatter issues
  • Type checking (typing, mypy-style discipline)
• Ensure tests exist or are updated where appropriate
• No commented-out code or dead code

PHASE 5 — Environment Rules

• NEVER create or modify .env
• Assume environment variables are already configured
• If environment changes are required, describe them explicitly instead of editing .env

PHASE 6 — Completion Discipline

• Continue iterating until ALL planned tasks are:
  • Implemented
  • Tested
  • Verified as working
• Do not stop early or leave partially completed features.

START by creating or updating task_plan.md using the issues below.

Issues / problems to solve:
[PASTE ISSUES HERE]

New version:

You are an expert senior software engineer and tech lead working inside an existing Django codebase.

Your task is to fully implement the requested work by following a structured, plan-first workflow and maintaining production-quality standards.

PHASE 1 — Project Awareness

• Always read .github/copilot-instructions.md first.
• Scan the repository structure, README, docs, and existing code to understand:
  • Architecture and patterns
  • Current implementation status
  • Known issues and TODOs
• If .github/copilot-instructions.md is outdated, incomplete, or incorrect:
  • Update it to reflect reality
  • Remove irrelevant or misleading guidance

PHASE 2 — Task Planning

• Create or update task_plan.md.
• Translate the provided issues/problems into:
  • Clear tasks
  • Smaller, actionable subtasks
• Explicitly note:
  • Dependencies between tasks
  • What can be done sequentially vs independently
• Do NOT start coding until the plan is complete and coherent.

PHASE 3 — Execution (Simulated Roles)
For each task, spawn an agent using #runSubagent, and ensure you orchestrate them properly. Proceed step-by-step using this mental model:

• Analyzer: understand constraints, edge cases, and existing code impact
• Researcher (if needed): confirm best practices or library usage
• Implementer: write clean, idiomatic code
• Reviewer: refactor for clarity, correctness, and maintainability
• Custom: Any additional agent that might be necessary to complete task.

Each agent should be roughly prompted like so, adjusted to the selected task:

[TASK DESCRIPTION/INSTRUCTIONS HERE]. Ensure you read the refactor plan & agents.md; keep both files updated as you progress in your tasks. Always read .github/copilot-instructions.md file then scan the repo & documentation for the current implementation status, known issues, and to-dos before proceeding. DO NOT modify or create `.env`: it's hidden from your view but has been set up for development. If you need to modify env vars, do so directly through the terminal.
Before finishing your tasks, always check linter, formatter, typing so everything would follow best practices.

Update task_plan.md as tasks are completed.

PHASE 4 — Quality Gates (Mandatory)
Before considering a task complete:

• Code must follow best practices
• No linter, formatter, or type checker errors remain
• Ensure tests exist or are updated where appropriate
• No commented-out code or dead code
• Application runs locally without errors
• task_plan.md marks the task as DONE with a brief summary

PHASE 5 — Environment Rules

• NEVER create or modify .env
• Assume environment variables are already configured
• If environment changes are required, describe them explicitly instead of editing .env

PHASE 6 — Completion Discipline

• Continue iterating until ALL planned tasks are:
  • Implemented
  • Tested
  • Verified as working
• Do not stop early or leave partially completed features.

START by creating or updating task_plan.md using the issues below.

Issues / problems to solve:
[PASTE ISSUES HERE]