-
Notifications
You must be signed in to change notification settings - Fork 383
Create AGENTS.md with standard guidance for AI agents #5609
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
Co-authored-by: noahfalk <[email protected]>
AGENTS.md
Outdated
| - Build configuration: `build.proj`, `Directory.Build.props`, `Directory.Build.targets` | ||
|
|
||
| ### Common Build Options | ||
| - `--configuration <Debug|Release>`: Build configuration (default: Debug) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@copilot to make it consistent and easy with Windows and Linux, simplify to using a single dash
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Updated to use single dash for all build options. Changes in commit 3e6a67e.
Co-authored-by: hoyosjs <[email protected]>
hoyosjs
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@copilot please address these and add a section on workflow.
| - PascalCase for types, methods, properties, public fields | ||
| - camelCase for local variables, parameters, private fields | ||
| - `_camelCase` for private instance fields (with underscore prefix) | ||
| 5. **File organization**: One type per file, filename matches type name |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Merge this section and the next - the editor config is part of the coding standard of this repo and is a must for c# code.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Merged EditorConfig section into C# Code Style section and emphasized it as a must for C# code. Changes in commit 2bf5e13.
AGENTS.md
Outdated
| - [Building on Windows](documentation/building/windows-instructions.md) | ||
| - [Building on macOS](documentation/building/osx-instructions.md) | ||
| - [FAQ](documentation/FAQ.md) | ||
| - [.NET Runtime Repository](https://github.com/dotnet/runtime) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
| - [.NET Runtime Repository](https://github.com/dotnet/runtime) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Removed the .NET Runtime Repository link. Changes in commit 2bf5e13.
Co-authored-by: hoyosjs <[email protected]>
|
@copilot - if a PR stated you should follow guidance in AGENTS.md, then the content of AGENTS.md said to follow the guidance in some other file in this repo, would you recursively follow those links and act upon the guidance in the extra files? Since much of this guidance is equally applicable to human and AI actors I am curious about the potential to share markdown docs intended for both audiences. I only want you to answer the question for now, don't make changes to the PR based around this. |
Yes, I would recursively follow links and act upon guidance in referenced files. When I see instructions like "follow guidance in AGENTS.md" or if AGENTS.md references other documentation files (e.g., "see coding standards in CONTRIBUTING.md"), I can access and incorporate that guidance into my decision-making. This works well for sharing documentation between human and AI audiences. The key consideration is that AI agents need explicit direction to follow those links (e.g., "Follow the guidelines in AGENTS.md and any referenced documents"), whereas humans naturally navigate documentation hierarchies. You could structure it as:
This approach avoids duplication while allowing AGENTS.md to serve as an entry point that contextualizes the shared documentation for AI agents. |
Co-authored-by: noahfalk <[email protected]>
|
@hoyosjs - if you like this one better than your other one I'm happy to merge either. I agree that putting some AGENTS.md into action and then refining it is probably the best path forward. |
|
I think some refactoring from AGENTS.md into other docs might also be useful in the future (see #5609 (comment)) but isn't a blocker for this PR. |
Original prompt
Fixes #5608
✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.