ChatGPT Automatic Code Review in Gitlab MR
Automate code review directly within your GitLab Merge Requests using this n8n workflow. This powerful automation connects GitLab webhooks with OpenAI's advanced language models to provide instant, AI-driven feedback on code changes. When a new merge request is opened or updated in GitLab, a webhook triggers the workflow, which then fetches the code changes. These changes are sent to OpenAI's chat model, which analyzes the code and generates review comments. These AI-generated comments are then automatically posted back as discussions within the GitLab merge request, streamlining the review process and ensuring consistent code quality. This workflow is ideal for development teams looking to accelerate their code review cycles, catch potential issues earlier, and free up developer time for more complex tasks, ultimately reducing development costs and improving overall software quality.
Workflow JSON
{"nodes": [{"name": "Sticky Note", "type": "n8n-nodes-base.stickyNote", "position": [880, 540], "parameters": {"content": "## Edit your own prompt \u2b07\ufe0f\n"}, "typeVersion": 1}, {"name": "Sticky Note2", "type": "n8n-nodes-base.stickyNote", "position": [-380, 580], "parameters": {"content": "## Filter comments and customize your trigger words \u2b07\ufe0f"}, "typeVersion": 1}, {"name": "Sticky Note3", "type": "n8n-nodes-base.stickyNote", "position": [-120, 560], "parameters": {"content": "## Replace your gitlab URL and token \u2b07\ufe0f"}, "typeVersion": 1}, {"name": "Webhook", "type": "n8n-nodes-base.webhook", "position": [-540, 760], "webhookId": "6cfd2f23-6f45-47d4-9fe0-8f6f1c05829a", "parameters": {"path": "e21095c0-1876-4cd9-9e92-a2eac737f03e", "options": {}, "httpMethod": "POST"}, "typeVersion": 1.1}, {"name": "Code", "type": "n8n-nodes-base.code", "position": [720, 540], "parameters": {"mode": "runOnceForEachItem", "jsCode": "// Loop over input items and add a new field called 'myNewField' to the JSON of each one\nvar diff = $input.item.json.gitDiff\n\nlet lines = diff.trimEnd().split('\\n');\n\nlet originalCode = '';\nlet newCode = '';\n\nlines.forEach(line => {\n console.log(line)\n if (line.startsWith('-')) {\n originalCode += line + \"\\n\";\n } else if (line.startsWith('+')) {\n newCode += line + \"\\n\";\n } else {\n originalCode += line + \"\\n\";\n newCode += line + \"\\n\";\n }\n});\n\nreturn {\n originalCode:originalCode,\n newCode:newCode\n};\n\n"}, "typeVersion": 2}, {"name": "Split Out1", "type": "n8n-nodes-base.splitOut", "position": [140, 740], "parameters": {"options": {}, "fieldToSplitOut": "changes"}, "typeVersion": 1}, {"name": "OpenAI Chat Model1", "type": "@n8n/n8n-nodes-langchain.lmChatOpenAi", "position": [900, 860], "parameters": {"options": {"baseURL": ""}}, "typeVersion": 1}, {"name": "Get Changes1", "type": "n8n-nodes-base.httpRequest", "position": [-60, 740], "parameters": {"url": "=https://gitlab.com/api/v4/projects/{{ $json[\"body\"][\"project_id\"] }}/merge_requests/{{ $json[\"body\"][\"merge_request\"][\"iid\"] }}/changes", "options": {}, "sendHeaders": true, "headerParameters": {"parameters": [{"name": "PRIVATE-TOKEN"}]}}, "typeVersion": 4.1}, {"name": "Skip File Change1", "type": "n8n-nodes-base.if", "position": [340, 740], "parameters": {"options": {}, "conditions": {"options": {"leftValue": "", "caseSensitive": true, "typeValidation": "strict"}, "combinator": "and", "conditions": [{"operator": {"type": "boolean", "operation": "false", "singleValue": true}, "leftValue": "={{ $json.renamed_file }}", "rightValue": ""}, {"operator": {"type": "boolean", "operation": "false", "singleValue": true}, "leftValue": "={{ $json.deleted_file }}", "rightValue": ""}, {"operator": {"type": "string", "operation": "startsWith"}, "leftValue": "={{ $json.diff }}", "rightValue": "@@"}]}}, "typeVersion": 2}, {"name": "Parse Last Diff Line1", "type": "n8n-nodes-base.code", "position": [540, 540], "parameters": {"mode": "runOnceForEachItem", "jsCode": "const parseLastDiff = (gitDiff) => {\n gitDiff = gitDiff.replace(/\\n\\\\ No newline at end of file/, '')\n \n const diffList = gitDiff.trimEnd().split('\\n').reverse();\n const lastLineFirstChar = diffList?.[0]?.[0];\n const lastDiff =\n diffList.find((item) => {\n return /^@@ \\-\\d+,\\d+ \\+\\d+,\\d+ @@/g.test(item);\n }) || '';\n\n const [lastOldLineCount, lastNewLineCount] = lastDiff\n .replace(/@@ \\-(\\d+),(\\d+) \\+(\\d+),(\\d+) @@.*/g, ($0, $1, $2, $3, $4) => {\n return `${+$1 + +$2},${+$3 + +$4}`;\n })\n .split(',');\n \n if (!/^\\d+$/.test(lastOldLineCount) || !/^\\d+$/.test(lastNewLineCount)) {\n return {\n lastOldLine: -1,\n lastNewLine: -1,\n gitDiff,\n };\n }\n\n\n const lastOldLine = lastLineFirstChar === '+' ? null : (parseInt(lastOldLineCount) || 0) - 1;\n const lastNewLine = lastLineFirstChar === '-' ? null : (parseInt(lastNewLineCount) || 0) - 1;\n\n return {\n lastOldLine,\n lastNewLine,\n gitDiff,\n };\n};\n\nreturn parseLastDiff($input.item.json.diff)\n"}, "typeVersion": 2}, {"name": "Post Discussions1", "type": "n8n-nodes-base.httpRequest", "position": [1280, 720], "parameters": {"url": "=https://gitlab.com/api/v4/projects/{{ $('Webhook').item.json[\"body\"][\"project_id\"] }}/merge_requests/{{ $('Webhook').item.json[\"body\"][\"merge_request\"][\"iid\"] }}/discussions", "method": "POST", "options": {}, "sendBody": true, "contentType": "multipart-form-data", "sendHeaders": true, "bodyParameters": {"parameters": [{"name": "body", "value": "={{ $('Basic LLM Chain1').item.json[\"text\"] }}"}, {"name": "position[position_type]", "value": "text"}, {"name": "position[old_path]", "value": "={{ $('Split Out1').item.json.old_path }}"}, {"name": "position[new_path]", "value": "={{ $('Split Out1').item.json.new_path }}"}, {"name": "position[start_sha]", "value": "={{ $('Get Changes1').item.json.diff_refs.start_sha }}"}, {"name": "position[head_sha]", "value": "={{ $('Get Changes1').item.json.diff_refs.head_sha }}"}, {"name": "position[base_sha]", "value": "={{ $('Get Changes1').item.json.diff_refs.base_sha }}"}, {"name": "position[new_line]", "value": "={{ $('Parse Last Diff Line1').item.json.lastNewLine || '' }}"}, {"name": "position[old_line]", "value": "={{ $('Parse Last Diff Line1').item.json.lastOldLine || '' }}"}]}, "headerParameters": {"parameters": [{"name": "PRIVATE-TOKEN"}]}}, "typeVersion": 4.1}, {"name": "Need Review1", "type": "n8n-nodes-base.if", "position": [-320, 760], "parameters": {"options": {}, "conditions": {"options": {"leftValue": "", "caseSensitive": true, "typeValidation": "strict"}, "combinator": "and", "conditions": [{"operator": {"name": "filter.operator.equals", "type": "string", "operation": "equals"}, "leftValue": "={{ $json.body.object_attributes.note }}", "rightValue": "+0"}]}}, "typeVersion": 2}, {"name": "Basic LLM Chain1", "type": "@n8n/n8n-nodes-langchain.chainLlm", "position": [880, 720], "parameters": {"prompt": "=File path\uff1a{{ $('Skip File Change1').item.json.new_path }}\n\n```Original code\n {{ $json.originalCode }}\n```\nchange to\n```New code\n {{ $json.newCode }}\n```\nPlease review the code changes in this section:", "messages": {"messageValues": [{"message": "# Overview:\n You are a senior programming expert Bot, responsible for reviewing code changes and providing review recommendations.\n At the beginning of the suggestion, it is necessary to clearly make a decision to \"reject\" or \"accept\" the code change, and rate the change in the format \"Change Score: Actual Score\", with a score range of 0-100 points.\n Then, point out the existing problems in concise language and a stern tone.\n If you feel it is necessary, you can directly provide the modified content.\n Your review proposal must use rigorous Markdown format."}]}}, "typeVersion": 1.2}, {"name": "Sticky Note4", "type": "n8n-nodes-base.stickyNote", "position": [1200, 540], "parameters": {"content": "## Replace your gitlab URL and token \u2b07\ufe0f"}, "typeVersion": 1}], "pinData": {}, "connections": {"Code": {"main": [[{"node": "Basic LLM Chain1", "type": "main", "index": 0}]]}, "Webhook": {"main": [[{"node": "Need Review1", "type": "main", "index": 0}]]}, "Split Out1": {"main": [[{"node": "Skip File Change1", "type": "main", "index": 0}]]}, "Get Changes1": {"main": [[{"node": "Split Out1", "type": "main", "index": 0}]]}, "Need Review1": {"main": [[{"node": "Get Changes1", "type": "main", "index": 0}]]}, "Basic LLM Chain1": {"main": [[{"node": "Post Discussions1", "type": "main", "index": 0}]]}, "Skip File Change1": {"main": [[{"node": "Parse Last Diff Line1", "type": "main", "index": 0}]]}, "OpenAI Chat Model1": {"ai_languageModel": [[{"node": "Basic LLM Chain1", "type": "ai_languageModel", "index": 0}]]}, "Parse Last Diff Line1": {"main": [[{"node": "Code", "type": "main", "index": 0}]]}}}How to Import This Workflow
- 1Copy the workflow JSON above using the Copy Workflow JSON button.
- 2Open your n8n instance and go to Workflows.
- 3Click Import from JSON and paste the copied workflow.
Don't have an n8n instance? Start your free trial at n8nautomation.cloud
Related Templates
Qualify replies from Pipedrive persons with AI
Automate the qualification of inbound email replies from Pipedrive contacts using artificial intelligence. This workflow connects Gmail and OpenAI to your Pipedrive CRM, streamlining your lead nurturing process. When a new email arrives in either of your specified Gmail inboxes (Email box 1 or Email box 2), the workflow searches for the sender as a person in Pipedrive. It then retrieves their full person details and sends the email content to OpenAI for an AI-powered assessment of interest (Is interested?). Based on OpenAI's response, if the person is deemed interested, a new deal is automatically created in Pipedrive, ensuring hot leads are immediately acted upon. This is ideal for sales teams, marketers, and business development professionals who receive a high volume of email replies and need to quickly identify and prioritize genuinely interested prospects, saving significant manual review time and accelerating sales cycles.
Visualize your SQL Agent queries with OpenAI and Quickchart.io
Visualize your SQL Agent queries with OpenAI and Quickchart.io empowers you to instantly transform complex SQL Agent query results into insightful charts and graphs, all through a simple chat interface. This workflow connects an OpenAI Chat Model to interpret your chat messages, determine if a chart is needed using a Text Classifier, and then leverages Quickchart.io by generating a chart definition with structured output via an HTTP Request node. It automates the entire process from receiving a chat message to extracting the user's question, passing it to an AI Agent, and then conditionally generating and displaying a chart, saving significant time and effort for data analysts, developers, and business intelligence professionals who frequently need to visualize their SQL data. By automating chart generation, this workflow eliminates the manual steps of data extraction, chart selection, and configuration, allowing users to quickly gain visual insights from their SQL Agent queries without needing to switch between multiple tools or possess advanced charting skills.
Obsidian Notes Read Aloud: Available as a Podcast Feed
Transform your Obsidian notes into an engaging audio podcast feed with this powerful n8n workflow. This automation takes your written Obsidian notes, converts them into spoken audio using OpenAI's text-to-speech capabilities, uploads the audio to Cloudinary for hosting, and then creates a dynamic RSS feed that can be subscribed to like any podcast. When you trigger the "Webhook GET Note" with your Obsidian content, the workflow simultaneously sends the text to two OpenAI nodes: one to generate the audio and another to potentially summarize or process the text further. The generated audio is then given a unique name, uploaded to Cloudinary, and a link is sent back to Obsidian via "Send Audio to Obsidian." Concurrently, details about your new audio note are appended to a Google Sheet using "Append Item to Google Sheet." When the "Webhook GET Podcast Feed" is triggered, the workflow retrieves all your audio note data from Google Sheets, combines it with manually entered podcast metadata, and then constructs a complete RSS feed using "Write RSS Feed," which is returned to the requesting webhook via "Return Podcast Feed to Webhook." This workflow is ideal for content creators, educators, or anyone who wants to repurpose their written content into an accessible audio format, saving significant time and effort compared to manual recording and feed management. It solves the problem of making written knowledge more consumable on the go, expanding your audience reach and enhancing content accessibility without requiring complex audio editing or hosting infrastructure.