⚙️ Review GitHub pull requests with GPT-4o and send feedback and Slack alerts
⚡ 131 views · ⚙️ DevOps & CI/CD
💡 Pro Tip — If you need GitHub data beyond what the REST API gives you, ScraperNode has a repository scraper that extracts metadata at scale without token rate limits.
Description
Automatically review pull requests with AI and post feedback as GitHub comments
Who is this for
Engineering teams, code reviewers, and tech leads who want to automate code review feedback. Ideal for repositories with high PR volume where consistent first-pass review is valuable.
What this workflow does
This workflow automates pull request code review using GPT-4o and routes findings to GitHub and Slack. When a pull request is opened, it fetches the raw diff, sends it to GPT-4o for analysis, and posts a formatted Markdown comment with severity rating, categorized issues, suggestions, and an overall score. Critical findings trigger immediate Slack alerts.
How to set up
- Add your GitHub OAuth2 credential to the trigger and comment nodes
- Add your OpenAI API credential to the GPT-4o node
- Add your Slack OAuth2 credential to both Slack notification nodes
- Configure n8n Variables:
GITHUB_OWNER,GITHUB_REPO, andGITHUB_TOKEN - Activate the workflow - the GitHub webhook registers automatically
- Open a test pull request to verify end-to-end execution
Requirements
- GitHub repository with OAuth2 credentials and Personal Access Token
- OpenAI API account with GPT-4o access
- Slack workspace with OAuth2 app installed
- Two Slack channels: one for critical alerts, one for review summaries
How to customize
Adjust the diff truncation limit (default 12,000 chars) in the Format PR Diff node. Modify the GPT-4o system prompt to focus on specific code quality concerns like security or performance. Update Slack channel routing based on your team’s notification policy.
Node List
| # | Node Name | Type | Purpose |
|---|---|---|---|
| 1 | When PR Opened | GitHub Trigger | Receives pull_request webhook events |
| 2 | Check PR Open | If | Passes only action = opened to avoid duplicates |
| 3 | Fetch PR Diff | HTTP Request | Fetches the raw unified diff from GitHub |
| 4 | Format PR Diff | Code | Extracts PR metadata; truncates diff to 12,000 chars |
| 5 | AI Code Review with GPT-4 | OpenAI | Sends diff to GPT-4o; returns structured JSON review |
| 6 | Build Comment for GitHub | Code | Parses JSON; assembles Markdown comment |
| 7 | Post Comment to GitHub | GitHub | Posts the AI review comment on the pull request |
| 8 | Check Critical Severity | If | Routes based on CRITICAL vs non-critical severity |
| 9 | Alert Critical Issues to Slack | Slack | Sends urgent alert to #incident |
| 10 | Share Review Summary on Slack | Slack | Sends summary notification to #code-reviews |
Total: 10 nodes (+ 5 Sticky Notes)
Sticky Note Compliance
| # | Sticky Note Title | Color | Role |
|---|---|---|---|
| 1 | Main Sticky Note (Overview) | Yellow | Workflow overview, How it works, Setup steps, Customization |
| 2 | Trigger and filter PR | White | Covers GitHub trigger and PR filter |
| 3 | Fetch and format diff | White | Covers diff retrieval and formatting |
| 4 | AI review and comment | White | Covers AI review and comment building |
| 5 | Severity check and alerts | White | Covers severity routing and Slack notifications |
All sticky notes use H2 headings (## ) and follow n8n public guidelines.
Tags
ai gpt-4 openai github slack code-review devops automation
🔗 Nodes Used
GitHub, Github Trigger, HTTP Request, Slack, OpenAI
📥 Import
Download workflow.json and import into n8n:
Workflow menu → Import from File