Skip to content
Closed
Changes from 1 commit
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
33 changes: 33 additions & 0 deletions rails/ai-rules/rails-rules.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,33 @@
# IDE AI thoughtbot rules

You are an expert in Ruby on Rails, PostgreSQL, and Hotwire (Turbo and Stimulus).

## Key Conventions

- Follow RESTful routing conventions: Seven restful actions: index, show, new, create, edit, update, delete (https://thoughtbot.com/blog/in-relentless-pursuit-of-rest-ish-routing)

Check failure on line 7 in rails/ai-rules/rails-rules.md

View workflow job for this annotation

GitHub Actions / lint

Bare URL used

rails/ai-rules/rails-rules.md:7:110 MD034/no-bare-urls Bare URL used [Context: "https://thoughtbot.com/blog/in..."] https://github.com/DavidAnson/markdownlint/blob/v0.37.4/doc/md034.md
- Use concerns for shared behavior across models or controllers
Comment thread
laicuRoot marked this conversation as resolved.
Outdated

## Data / Models

- To find model structure look in `db/schema.rb`
- When working with model attributes don’t guess, grep the schema at `db/schema.rb` to confirm and use only valid attributes

## UI and Styling

- Use Rails view helpers and partials to keep views DRY

## Performance Optimization

- Always check for N+1 queries when rendering collections
Copy link
Copy Markdown
Contributor

@louis-antonopoulos louis-antonopoulos Jan 9, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I would remove or change this line. I think it's more beneficial to use a tool like Bullet or Rails' strict_loading and rely on tests failing because of N+1 violations, which then guide the agent to fixing the issue.

I say this because there's a balance between telling the agent too little and too much, and I see this instruction as an opportunity cost:

"What might the agent not listen to because it was focused on this instruction, an issue that we have simple ways to check for?"

So while I would remove it, if kept, maybe something like

Suggested change
- Always check for N+1 queries when rendering collections
- Avoid N+1 queries when rendering collections

- Prefer includes for eager loading
- Scope queries to only the fields needed with select
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Does this mean that it will add .select to all queries? While there are performance benefits on that, I'd say it's premature optimization most of the time.


## Testing

- Always write tests to cover new code generated
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Should we tell it to write tests first? I think @JoelQ had more success with that.

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I have struggled with this one personally when working with Cursor. Normally I will write some basic tests, and based on that, I would get a much better suggestion for the feature and then iterate. After having the firsts tests, the tests examples that come after are good.

@JoelQ do you have any suggestions about how to make the rules effective to follow TDD?

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I've usually done that with an initial prompt. Zed's agent (using Claude) does this well.

- Use RSpec for testing framework
Comment thread
laicuRoot marked this conversation as resolved.
Outdated
- Use factories (FactoryBot) (https://thoughtbot.github.io/factory_bot/)

Check failure on line 29 in rails/ai-rules/rails-rules.md

View workflow job for this annotation

GitHub Actions / lint

Bare URL used

rails/ai-rules/rails-rules.md:29:31 MD034/no-bare-urls Bare URL used [Context: "https://thoughtbot.github.io/f..."] https://github.com/DavidAnson/markdownlint/blob/v0.37.4/doc/md034.md
- In tests, avoid lets and before (avoid mystery guests), do test setup within each test
- Verify new code by running test files using `bundle exec rspec spec/path/to/file_spec.rb`
- You can run a specific test by appending the line number (it can be any line number starting from the "it" block of the test) eg. `bundle exec rspec spec/path/to/file_spec.rb:72`

Loading