-
Notifications
You must be signed in to change notification settings - Fork 5.3k
Split Copilot instructions into agent-specific files #122413
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: stephentoub <2642209+stephentoub@users.noreply.github.com>
Co-authored-by: stephentoub <2642209+stephentoub@users.noreply.github.com>
Co-authored-by: stephentoub <2642209+stephentoub@users.noreply.github.com>
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.
Pull request overview
This PR successfully splits the Copilot instructions into two agent-specific files following GitHub's new feature for targeting different AI agents. The coding agent will focus on implementing changes correctly, while the code review agent will focus on expert-level review concerns.
Key Changes:
- Added YAML frontmatter to both files to exclude the non-relevant agent
- Kept existing comprehensive build/test/coding guidelines in the coding agent instructions
- Created focused code review instructions emphasizing security, performance, compatibility, and design over style
Reviewed changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.
| File | Description |
|---|---|
.github/copilot-instructions.md |
Added YAML frontmatter with excludeAgent: code-review-agent to target only the coding agent, keeping all existing build, test, and coding style guidelines |
.github/code-review-instructions.md |
New file with excludeAgent: coding-agent targeting the code review agent, containing comprehensive review priorities for security, performance, backwards compatibility, cross-component interactions, correctness, design, testing, and documentation |
The changes are well-structured and align with the repository's needs. The code review instructions appropriately focus on expert judgment areas (security, performance, compatibility) while explicitly excluding automated tooling concerns (formatting, syntax). Both files follow proper YAML frontmatter syntax and markdown formatting without trailing whitespace issues.
|
/ba-g markdown only |
jkotas
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.
Is it possible to see examples code-review feedback improved by this prompt? Or is the "production" only reasonable way to see how this prompt is going to work in practive?
Unfortunately, not to my knowledge. I've shared such feedback that there needs to be a tighter inner loop on iterating on these. |
|
/ba-g markdown only |
Plan: Split Copilot Instructions for Coding Agent and Code Review Agent
.github/copilot-instructions.md(coding agent instructions) with existing contentexcludeAgent: code-review-agentfrontmatter.github/code-review-instructions.md(code review agent instructions) focused on:excludeAgent: coding-agentfrontmatterexcludeAgent: codingtoexcludeAgent: coding-agentexcludeAgent: code-reviewtoexcludeAgent: code-review-agentSummary
Successfully split the Copilot instructions into two specialized files with proper agent exclusion:
.github/copilot-instructions.md- Coding Agent Instructions (excludes code-review-agent)excludeAgent: code-review-agent.github/code-review-instructions.md- Code Review Agent Instructions (excludes coding-agent)excludeAgent: coding-agentEach agent now only sees the instructions relevant to its role.
Original prompt
💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.