- You are a thorough researcher.
- You use tools like web search and context7 to find documentation and answers.
- You always think about several alternatives to solve a problem and list the option that you recommend the most last.

Do some research on the following topic:

<research_topic>
$ARGUMENTS
</research_topic>

Draft a detailed, step-by-step blueprint for building this feature:

<feature>
  $ARGUMENTS
</feature>

Then, once you have a solid plan, break it down into small, iterative chunks that build on each other. Look at these chunks and then go another round to break it into small steps. review the results and make sure that the steps are small enough to be implemented safely, but big enough to move the project forward. Iterate until you feel that the steps are right sized for this project.

From here you should have the foundation to provide a series of prompts for a code-generation LLM that will implement each step. Prioritize best practices, and incremental progress, ensuring no big jumps in complexity at any stage. Make sure that each prompt builds on the previous prompts, and ends with wiring things together. There should be no hanging or orphaned code that isn't integrated into a previous step.

Make sure and separate each prompt section. Use markdown. Each prompt should be tagged as text using code tags. The goal is to output prompts, but context, etc is important as well.

IMPORTANT: focus very hard on the task at hand. Plan only for the task at hand, do not include any unsolicitated UX improvements, and keep the UI design minimal.

Add only tests when asked to, but make sure the code is testable.

Draft a detailed, step-by-step blueprint for building this project. Then, once you have a solid plan, break it down into small, iterative chunks that build on each other. Look at these chunks and then go another round to break it into small steps. review the results and make sure that the steps are small enough to be implemented safely, but big enough to move the project forward. Iterate until you feel that the steps are right sized for this project.

From here you should have the foundation to provide a series of prompts for a code-generation LLM that will implement each step. Prioritize best practices, and incremental progress, ensuring no big jumps in complexity at any stage. Make sure that each prompt builds on the previous prompts, and ends with wiring things together. There should be no hanging or orphaned code that isn't integrated into a previous step.

Make sure and separate each prompt section. Use markdown. Each prompt should be tagged as text using code tags. The goal is to output prompts, but context, etc is important as well.

Store the plan in plan.md. Also create a todo.md to keep state.

The spec is in the file called: $ARGUMENTS

Create `{timestamp}-{slug}.md` with a complete summary of our research session.
Save it inside the `.claude/research` directory in the project.

Here are some last instructions that you need to account for: $ARGUMENTS

Include:

- URLs to resources as footnotes, e.g.:

<markdown>
Tis is an interesting statement about sharks[^1].

[^1]:
    Sharks suffocate when they don't move, according to [Science Journal](https://example.com)
    </markdown>

npx ccusage@latest