⚙️ Convert GitHub commits into review-ready pull requests with Google Gemini
⚡ 23 views · ⚙️ DevOps & CI/CD
Description
Description
Categories
Developer Automation, AI Agents, GitHub Automation, DevOps Productivity
Build an AI-Driven GitHub Pull Request Automation with n8n + MCP
This workflow creates an AI-powered GitHub automation that turns raw commit history into a clean, professional pull request automatically.
When triggered via MCP or another workflow, it extracts repository details, fetches all commits from a target branch, uses AI to understand the intent behind the changes, and creates a well-structured pull request with a clear title and description.
The result is a reliable, no-manual-work system that standardizes pull requests and reduces review friction across teams.
Benefits
Consistent Pull Requests
Every PR follows a clean, readable structure regardless of who triggered it.
Zero Manual Formatting
No copy-pasting commit messages or writing descriptions by hand.
Faster Review Cycles
Reviewers get clear context upfront, reducing back-and-forth.
AI-Assisted Context Awareness
Commit history is summarized intelligently, not blindly concatenated.
MCP-Ready Automation
Can be called directly by AI tools like Cursor through MCP.
How It Works
MCP or Workflow Trigger
- Triggered via MCP server or another n8n workflow
- Accepts natural language or structured input
Repository Information Extraction
- AI extracts:
- Repository owner
- Repository name
- Source branch
- Base branch
Commit Retrieval (GitHub API)
- Fetches all commits for the source branch
- Collects commit messages as context
Commit Summarization (AI)
- AI analyzes commit history
- Generates:
- A concise PR title
- A clear bullet-point description
Pull Request Creation
- Creates a GitHub pull request automatically
- Uses correct base and head branches
- Inserts AI-generated title and description
Required Setup
GitHub
- Repository access
- OAuth or personal access token
- Permission to read commits and create pull requests
AI Model
- Google Gemini or compatible LLM
- Connected via n8n AI nodes
n8n
- Self-hosted or cloud
- HTTP access to GitHub APIs
- MCP Trigger enabled for AI tool access
Business Use Cases
Engineering Teams
- Standardize PR quality across developers
- Reduce cognitive load on contributors
DevOps & Platform Teams
- Enforce PR hygiene automatically
- Improve velocity without extra process
Founders & Tech Leads
- Maintain clean repositories without micromanagement
Agencies & Consultants
- Deliver AI-assisted GitHub automation to clients
Difficulty Level
Intermediate
Estimated Build Time
45–75 minutes
Monthly Operating Cost
- GitHub: Existing plan
- AI Model: Free tier or usage-based
- n8n: Self-hosted or cloud
Typical range: $0–20/month
Why This Workflow Works
- Commits are the most reliable source of change intent
- AI summarizes meaning, not noise
- MCP enables direct AI-to-automation execution
- GitHub remains the single source of truth
Possible Extensions
- Auto-assign reviewers based on files changed
- Add PR labels using AI classification
- Generate changelog entries automatically
- Post PR summary to Slack or Teams
- Enforce branch naming or commit standards
Details
Nodes used in workflow
- MCP Trigger
- AI Agent (Repository Extraction)
- Structured Output Parser
- GitHub API (Commits)
- Summarize
- AI LLM Chain
- GitHub API (Create Pull Request)
- If
- Edit Fields (Set)
- Sticky Note
🔗 Nodes Used
HTTP Request, Execute Workflow Trigger, AI Agent, Basic LLM Chain, Structured Output Parser, Call n8n Workflow Tool
📥 Import
Download workflow.json and import into n8n:
Workflow menu → Import from File