coderabbit-core-workflow-a
Configures automated PR reviews and feedback loops using CodeRabbit's core workflow.
Install
mkdir -p .claude/skills/coderabbit-core-workflow-a && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/5445" && unzip -o skill.zip -d .claude/skills/coderabbit-core-workflow-a && rm skill.zipInstalls to .claude/skills/coderabbit-core-workflow-a
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Execute CodeRabbit primary workflow: automated PR code review with configuration.Key capabilities
- →Configure review profiles for automated PR analysis
- →Generate walkthrough summaries and line-level comments
- →Enable or disable auto-review for draft PRs
- →Trigger specific commands like 'full review' or 'generate-docstrings'
- →Configure finishing touch recipes for automated code improvements
How it works
CodeRabbit analyzes the diff of a pull request based on the `.coderabbit.yaml` configuration, then posts a walkthrough summary and line-level comments. Developers address feedback, and CodeRabbit performs incremental re-reviews on new commits.
Inputs & outputs
When to use coderabbit-core-workflow-a
- →Setting up automated PR reviews
- →Configuring review feedback profiles
- →Establishing PR review workflows
- →Managing CodeRabbit review status
About this skill
CodeRabbit Core Workflow A: Automated PR Review
Overview
The primary CodeRabbit workflow: a developer opens a PR, CodeRabbit automatically analyzes the diff, posts a walkthrough summary and line-level comments, and the developer addresses feedback. This skill covers configuration, review profiles, path instructions, and the full review lifecycle.
Prerequisites
- CodeRabbit GitHub App installed (see
coderabbit-install-auth) .coderabbit.yamlin repository root- At least one PR-capable branch
Instructions
Step 1: Configure the Review Pipeline
# .coderabbit.yaml - Production-ready configuration
language: "en-US"
early_access: false
reviews:
profile: "assertive" # chill = less feedback, assertive = more thorough
request_changes_workflow: true # CodeRabbit marks review as "changes requested" for issues
high_level_summary: true # Post a walkthrough comment summarizing all changes
high_level_summary_in_walkthrough: true
review_status: true # Show review progress status
collapse_walkthrough: false # Keep walkthrough expanded
sequence_diagrams: true # Generate control flow diagrams
poem: false # Disable poems in review summary
auto_review:
enabled: true
drafts: false # Skip draft PRs
base_branches:
- main
- develop
ignore_title_keywords:
- "WIP"
- "DO NOT MERGE"
- "chore: bump"
path_filters:
- "!**/*.lock"
- "!**/*.snap"
- "!**/generated/**"
- "!dist/**"
- "!**/*.min.js"
- "!vendor/**"
path_instructions:
- path: "src/api/**"
instructions: |
Review for: input validation, proper HTTP status codes, auth middleware usage,
error response format per RFC 7807. Flag missing error handling.
- path: "src/db/**"
instructions: |
Review for: parameterized queries (no string concatenation), transaction boundaries,
proper connection cleanup, index usage. Flag N+1 query patterns.
- path: "**/*.test.*"
instructions: |
Review for: assertion completeness, edge case coverage, proper async handling.
Do NOT comment on test naming conventions or import order.
- path: ".github/workflows/**"
instructions: |
Review for: pinned action versions (use SHA not tag), no secrets in logs,
timeout-minutes on all jobs, OIDC for cloud auth.
chat:
auto_reply: true
# Finishing touches configuration
reviews:
finishing_touches:
docstrings:
enabled: true # Allow @coderabbitai generate-docstrings command
Step 2: Understand the Review Lifecycle
Developer opens/updates PR
│
▼
┌─────────────────────────────────┐
│ CodeRabbit analyzes diff │
│ (typically 2-5 min, up to 15 │
│ min for 1000+ line PRs) │
└─────────┬───────────────────────┘
│
├──▶ Walkthrough comment (summary + sequence diagram)
│
├──▶ Line-level comments (bugs, suggestions, improvements)
│
└──▶ Review state (approved / changes_requested)
│
▼
Developer addresses feedback
│
┌─────────┴──────────┐
│ │
Reply to comment Push new commits
(conversation) (incremental re-review)
│ │
▼ ▼
CodeRabbit responds CodeRabbit reviews
with explanation only changed files
Step 3: Interact with Reviews
# In any PR comment:
@coderabbitai full review # Re-review all files from scratch
@coderabbitai summary # Regenerate walkthrough summary
@coderabbitai resolve # Mark all CodeRabbit comments as resolved
@coderabbitai generate-docstrings # Auto-generate docstrings for functions
@coderabbitai configuration # Show current active config as YAML
@coderabbitai help # List all commands
# Reply to any CodeRabbit inline comment to discuss the feedback.
# CodeRabbit maintains conversation context and will explain its reasoning.
# In PR description, add instructions for this specific review:
# "Focus on security implications of the auth changes"
Step 4: Configure Finishing Touch Recipes
# .coderabbit.yaml - Custom finishing touch recipes (open beta)
finishing_touches:
recipes:
- name: "fix-imports"
description: "Sort and organize imports"
instructions: |
Sort all imports alphabetically. Group: external packages first,
then internal modules, then relative imports. Remove unused imports.
- name: "tighten-types"
description: "Replace any with specific types"
instructions: |
Replace all `any` types with proper TypeScript types.
Use `unknown` for truly unknown values. Add type guards where needed.
# Trigger recipes in a PR comment:
@coderabbitai run fix-imports
@coderabbitai run tighten-types
# Or check the boxes in the Finishing Touches section of the walkthrough
Output
- Automated review on every PR targeting configured branches
- Walkthrough summary with sequence diagrams
- Line-level feedback categorized by severity
- Interactive conversation on review comments
- Finishing touch recipes for automated code improvements
Error Handling
| Issue | Cause | Solution |
|---|---|---|
| Review takes 15+ minutes | PR has 1000+ changed lines | Split into smaller PRs |
| No review posted | PR targets non-configured branch | Add branch to base_branches |
| Reviews on generated files | Missing path_filters | Add !**/generated/** to path_filters |
| Too many nitpick comments | Profile set to assertive | Switch to chill for experienced teams |
| Config changes not applied | YAML syntax error | Run @coderabbitai configuration to verify |
| Review on draft PR | drafts: true in config | Set drafts: false to skip drafts |
Resources
Next Steps
For configuration tuning and noise reduction, see coderabbit-core-workflow-b.
When not to use it
- →When the PR targets a non-configured branch
- →When the `.coderabbit.yaml` file has a syntax error
- →When the PR has over 1000 changed lines and a quick review is needed
Prerequisites
Limitations
- →Reviews may take up to 15 minutes for PRs with over 1000 changed lines
- →Reviews are not posted if the PR targets a non-configured branch
- →Generated files may be reviewed if `path_filters` are missing
How it compares
This workflow automates code review and feedback generation directly within pull requests, unlike manual review processes that require human intervention for initial analysis and comment generation.
Compared to similar skills
coderabbit-core-workflow-a side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| coderabbit-core-workflow-a (this skill) | 1 | 27d | Review | Intermediate |
| resolve-conflicts | 81 | 8mo | Review | Intermediate |
| claude-automation-recommender | 47 | 2mo | Review | Beginner |
| codex-skill | 12 | 5mo | Review | Advanced |
Try saying
Example prompts that trigger this skill in your AI assistant.
More by jeremylongshore
View all by jeremylongshore →You might also like
resolve-conflicts
antinomyhq
Use this skill immediately when the user mentions merge conflicts that need to be resolved. Do not attempt to resolve conflicts directly - invoke this skill first. This skill specializes in providing a structured framework for merging imports, tests, lock files (regeneration), configuration files, and handling deleted-but-modified files with backup and analysis.
claude-automation-recommender
anthropics
Analyze a codebase and recommend Claude Code automations (hooks, subagents, skills, plugins, MCP servers). Use when user asks for automation recommendations, wants to optimize their Claude Code setup, mentions improving Claude Code workflows, asks how to first set up Claude Code for a project, or wants to know what Claude Code features they should use.
codex-skill
feiskyer
Use when user asks to leverage codex, gpt-5, or gpt-5.1 to implement something (usually implement a plan or feature designed by Claude). Provides non-interactive automation mode for hands-off task execution without approval prompts.
git-advanced-workflows
wshobson
Master advanced Git workflows including rebasing, cherry-picking, bisect, worktrees, and reflog to maintain clean history and recover from any situation. Use when managing complex Git histories, collaborating on feature branches, or troubleshooting repository issues.
subagent-driven-development
davila7
Use when executing implementation plans with independent tasks in the current session
validate-openapi-specs
epieczko
Validates and registers hook manifest files (YAML) in the Hook Registry for versioned hook management.