Add agent instruction documentation for AI-assisted development #976
Conversation
Documents standards for AI code generation and review protocols including mandatory review markers, investigation procedures, and project-specific development workflows. Signed-off-by: Gorkem Ercan <[email protected]>
There was a problem hiding this comment.
Pull Request Overview
Adds comprehensive documentation for AI-assisted development standards, including mandatory review protocols for AI-generated code. This establishes quality gates and accountability measures for contributions that use AI coding tools while maintaining project code quality standards.
- Introduces AI contribution guidelines with mandatory review requirements and testing standards
- Implements agent review markers system to prevent unreviewed AI code from merging
- Adds structured instructions for both code generation and review processes
Reviewed Changes
Copilot reviewed 4 out of 4 changed files in this pull request and generated 3 comments.
| File | Description |
|---|---|
| CONTRIBUTING.md | Adds AI-assisted contributions section with requirements and reviewer guidelines |
| CLAUDE.md | Creates redirect file pointing to AGENTS.md |
| AGENTS.md | Comprehensive agent instructions covering code generation, review protocols, and project architecture |
| .github/pull_request_template.md | Adds AI-assisted code checklist to PR template |
Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.
| @@ -0,0 +1 @@ | |||
| AGENTS.md No newline at end of file | |||
There was a problem hiding this comment.
This file appears to be a redirect but lacks proper documentation format. Consider adding a proper header and explanation, such as '# Claude Agent Instructions\n\nSee AGENTS.md for comprehensive agent instructions and development guidelines.'
| AGENTS.md | |
| # Claude Agent Instructions | |
| See [AGENTS.md](AGENTS.md) for comprehensive agent instructions and development guidelines. |
Copilot uses AI. Check for mistakes.
| **What you MUST ignore:** | ||
|
|
||
| - Urges to refactor adjacent code | ||
| - Desire to add "helpful" documentation | ||
| - Temptation to implement "better" solutions than specified | ||
| - Any scope creep whatsoever |
There was a problem hiding this comment.
[nitpick] The formatting inconsistency between numbered list items (1-6) and bullet points could be confusing. Consider using consistent formatting throughout the section, either all numbered items or all bullet points.
Copilot uses AI. Check for mistakes.
| ## Project Overview | ||
|
|
There was a problem hiding this comment.
This project description refers to KitOps, but the earlier sections (lines 205-345) describe the Jozu Hub product repository. This creates confusion about which project this documentation is for. Consider clarifying the relationship between these projects or separating the documentation appropriately.
| ## Project Overview | |
| ## Relationship Between Jozu Hub and KitOps | |
| Jozu Hub integrates KitOps as its reference implementation for packaging, versioning, and securely sharing AI/ML projects. KitOps provides OCI-compliant model packaging and interchange, which Jozu Hub leverages to ensure vendor-neutral AI/ML workflows. | |
| ## KitOps Project Overview |
Copilot uses AI. Check for mistakes.
Documents standards for AI code generation and review protocols including mandatory review markers, investigation procedures, and project-specific development workflows.