[Feat] Add agent guidance and code review skill - #1211
Conversation
📝 WalkthroughWalkthroughAdds a code-review skill with parallel Specs and Standards checks, and changes ChangesCode review setup
Estimated code review effort: 2 (Simple) | ~10 minutes 🚥 Pre-merge checks | ✅ 4✅ Passed checks (4 passed)
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In @.agents/skills/code-review/SKILL.md:
- Around line 18-24: Update the Review Process in SKILL.md to explicitly define
whether this two-agent workflow complements or replaces the existing /review-pr
contract. Preserve /review-pr’s three specialised reviewer coverage and
prioritised findings, and describe how these agents’ results integrate without
removing or weakening that behaviour.
- Around line 14-16: Update the Standards section in the review skill to
explicitly require inspecting the authoritative repository instructions,
including .github/copilot-instructions.md and all applicable
.github/instructions/*.md files, alongside AGENTS.md or CLAUDE.md. Ensure
standards reviews are based on these sources rather than only the latter files.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro Plus
Run ID: 03c9c68f-156f-4c6f-bf50-90652cfa7025
📒 Files selected for processing (3)
.agents/skills/code-review/SKILL.md.gitignoreAGENTS.md
💤 Files with no reviewable changes (1)
- .gitignore
| ## Standards | ||
|
|
||
| Understand the project standards, check existing patterns, read AGENTS.md or CLAUDE.md. |
There was a problem hiding this comment.
🎯 Functional Correctness | 🟠 Major | ⚡ Quick win
Read the authoritative repository instructions, not only AGENTS.md or CLAUDE.md.
CLAUDE.md states that the authoritative standards are in .github/copilot-instructions.md and .github/instructions/*.md. Without explicitly requiring the Standards Agent to inspect those files, this skill can report a standards review based on incomplete guidance.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In @.agents/skills/code-review/SKILL.md around lines 14 - 16, Update the
Standards section in the review skill to explicitly require inspecting the
authoritative repository instructions, including .github/copilot-instructions.md
and all applicable .github/instructions/*.md files, alongside AGENTS.md or
CLAUDE.md. Ensure standards reviews are based on these sources rather than only
the latter files.
| ## Review Process | ||
|
|
||
| Spawn 2 independent and parallel sub-agents without sharing or accessing context to eachother. | ||
|
|
||
| **Specs Agent:** Should review the code changes to verify if they are achieving the specs or not. | ||
|
|
||
| **Standards Agent:** Should review the code changes to verify if they are following the project standards or not. |
There was a problem hiding this comment.
🎯 Functional Correctness | 🟠 Major | ⚡ Quick win
Reconcile this workflow with the existing /review-pr contract.
CLAUDE.md documents /review-pr as running three specialised reviewer agents and returning prioritised findings, whereas this skill mandates two generic agents and does not define how their results integrate with that workflow. Please state whether this skill complements or replaces /review-pr, and preserve the existing reviewer coverage and prioritisation behaviour where applicable.
🧰 Tools
🪛 LanguageTool
[style] ~22-~22: In contexts where ‘if’ is followed by ‘or’, using ‘whether’ may be more appropriate (and formal).
Context: ...hould review the code changes to verify if they are achieving the specs or not. *...
(IF_WHETHER)
[style] ~24-~24: In contexts where ‘if’ is followed by ‘or’, using ‘whether’ may be more appropriate (and formal).
Context: ...hould review the code changes to verify if they are following the project standard...
(IF_WHETHER)
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In @.agents/skills/code-review/SKILL.md around lines 18 - 24, Update the Review
Process in SKILL.md to explicitly define whether this two-agent workflow
complements or replaces the existing /review-pr contract. Preserve /review-pr’s
three specialised reviewer coverage and prioritised findings, and describe how
these agents’ results integrate without removing or weakening that behaviour.
saeedvaziry
changed the title
1 participant
Track AGENTS.md as a symlink to CLAUDE.md and add a code-review skill that independently evaluates changes against specifications and project standards.
Summary by CodeRabbit