mirror of
https://github.com/khoaliber/n8nworkflows.xyz.git
synced 2026-04-19 17:14:37 +00:00
creation
This commit is contained in:
nusquama
@@ -0,0 +1,129 @@
|
||||
Generate, review, and optimize code with Cursor AI, GitHub, Google, and Slack
|
||||
|
||||
https://n8nworkflows.xyz/workflows/generate--review--and-optimize-code-with-cursor-ai--github--google--and-slack-13827
|
||||
|
||||
|
||||
# Generate, review, and optimize code with Cursor AI, GitHub, Google, and Slack
|
||||
|
||||
# Reference Document: n8n Cursor AI Coding Workflow
|
||||
|
||||
## 1. Workflow Overview
|
||||
The **n8n Cursor AI Coding Workflow** is an automated pipeline designed to bridge the gap between high-level task descriptions and production-ready code. It leverages the advanced coding intelligence of Cursor AI to automate the entire lifecycle of a coding task: generation, review, optimization, and deployment.
|
||||
|
||||
By integrating directly with GitHub, Google Workspace, and Slack, this workflow transforms n8n into a continuous integration/continuous delivery (CI/CD) engine for AI-assisted development. It ensures that every piece of AI-generated code meets a specific quality threshold before being committed, documented, and logged.
|
||||
|
||||
### Logical Blocks
|
||||
1. **1.1 Input Reception & Configuration:** Captures incoming tasks via Webhook or Schedule and sets the operational environment variables.
|
||||
2. **1.2 Intelligence Phase (Classification):** Uses LLMs to categorize the request (e.g., Generate vs. Refactor) to select the correct prompt strategy.
|
||||
3. **1.3 Execution Phase (Cursor AI):** Calls the Cursor AI API to generate the code and subsequently perform a structured code review.
|
||||
4. **1.4 Quality Gate:** Evaluates the AI's self-review score against a user-defined threshold.
|
||||
5. **1.5 Deployment & Notification:** Commits approved code to GitHub, saves reports to Google Drive, logs data to Sheets, and alerts the team via Slack.
|
||||
|
||||
---
|
||||
|
||||
## 2. Block-by-Block Analysis
|
||||
|
||||
### 2.1 Trigger & Task Input
|
||||
- **Overview:** Receives the coding task and initializes the workflow with repository settings, quality thresholds, and model selections.
|
||||
- **Nodes Involved:** `Webhook: Receive Coding Task`, `Daily Batch (fallback)`, `Set Task Config`.
|
||||
- **Node Details:**
|
||||
- **Webhook / Schedule Trigger:** Accepts a POST request with a `task` body or triggers every 24 hours.
|
||||
- **Set Task Config (Set Node):**
|
||||
- Defines hardcoded variables: `repo_owner`, `repo_name`, `target_branch`, `language` (default: python), `quality_threshold` (default: 75), `storage_folder`, `log_sheet_id`, and `cursor_model` (cursor-fast).
|
||||
- *Edge Case:* If the JSON body in the webhook is missing the `task` key, subsequent AI nodes will fail.
|
||||
|
||||
### 2.2 Classification Phase
|
||||
- **Overview:** Analyzes the user's prompt to determine the intent and technical requirements.
|
||||
- **Nodes Involved:** `Create a conversation`, `Wait: Processing Delay`, `Parse Classification`.
|
||||
- **Node Details:**
|
||||
- **Create a conversation (OpenAI):** Uses OpenAI to interpret the task and output a classification.
|
||||
- **Wait (Wait Node):** A short delay to ensure API stability and avoid rate limits.
|
||||
- **Parse Classification (Code Node):**
|
||||
- Uses regex (`/```json|```/g`) to clean the AI's response and parse it into `task_type`, `language`, and `summary`.
|
||||
- *Failure Type:* Will error if the AI output is not valid JSON or contains unexpected text.
|
||||
|
||||
### 2.3 Cursor Phase
|
||||
- **Overview:** The core processing engine where the code is authored and critiqued.
|
||||
- **Nodes Involved:** `Cursor AI: Generate Code`, `Cursor AI: Code Review`.
|
||||
- **Node Details:**
|
||||
- **Cursor AI: Generate Code (HTTP Request):**
|
||||
- Sends a System Prompt defining the AI as an expert developer.
|
||||
- Parameters: `model` (from config), `max_tokens` (2000).
|
||||
- Requires: `cursorApiKey`.
|
||||
- **Cursor AI: Code Review (HTTP Request):**
|
||||
- Sends the generated code back to Cursor for a "Senior Reviewer" audit.
|
||||
- Instructions: Must return JSON with `score`, `issues`, and `suggestions`.
|
||||
|
||||
### 2.4 Quality Gate
|
||||
- **Overview:** A logic check that decides whether the code is ready for production or requires an optimization pass.
|
||||
- **Nodes Involved:** `Quality Score OK?`.
|
||||
- **Node Details:**
|
||||
- **Quality Score OK? (If Node):** Compares the `score` from the review node against the `quality_threshold` (e.g., 75) defined in the Config node.
|
||||
|
||||
### 2.5 Output Phase (Deployment & Logging)
|
||||
- **Overview:** Finalizes the task by pushing to Git and updating management tools.
|
||||
- **Nodes Involved:** `GitHub: Commit Code`, `Cursor AI: Optimize Code`, `Google Drive: Save Report`, `Google Sheets: Log Result`, `Notify Slack: Success`, `Notify Slack: Optimized`.
|
||||
- **Node Details:**
|
||||
- **GitHub: Commit Code (HTTP Request):** Base64 encodes the code and pushes it to GitHub using the GitHub REST API. Dynamic filename based on timestamp.
|
||||
- **Google Drive: Save Report:** Creates a Markdown file containing the review details in the specified folder.
|
||||
- **Google Sheets: Log Result:** Appends a row with the Task, Status, Language, Score, and GitHub link.
|
||||
- **Slack Nodes:** Sends formatted messages to the `ai-code-review` channel.
|
||||
|
||||
---
|
||||
|
||||
## 3. Summary Table
|
||||
|
||||
| Node Name | Node Type | Functional Role | Input Node(s) | Output Node(s) | Sticky Note |
|
||||
| :--- | :--- | :--- | :--- | :--- | :--- |
|
||||
| Webhook: Receive Coding Task | Webhook | Entry Point | - | Set Task Config | 1. Trigger & Task Input |
|
||||
| Daily Batch (fallback) | Schedule | Entry Point | - | Set Task Config | 1. Trigger & Task Input |
|
||||
| Set Task Config | Set | Variable Init | Webhook, Schedule | Create a conversation | 1. Trigger & Task Input |
|
||||
| Create a conversation | OpenAI | AI Intent Parsing | Set Task Config | Wait | 2. Task Classification |
|
||||
| Wait: Processing Delay | Wait | Flow Control | Create a conversation | Parse Classification | 2. Task Classification |
|
||||
| Parse Classification | Code | Data Cleanup | Wait | Cursor AI: Generate Code | 2. Task Classification |
|
||||
| Cursor AI: Generate Code | HTTP Request | Code Writing | Parse Classification | Cursor AI: Code Review | 3. Cursor AI Code Generation |
|
||||
| Cursor AI: Code Review | HTTP Request | Quality Audit | Cursor AI: Gen Code | Quality Score OK? | 3. Cursor AI Code Generation |
|
||||
| Quality Score OK? | If | Logic Gate | Cursor AI: Review | GitHub, Cursor AI: Opt | 4. Quality Gate |
|
||||
| Cursor AI: Optimize Code | HTTP Request | Self-Correction | Quality Score OK? | Notify Slack: Optimized | 3. Cursor AI Code Generation |
|
||||
| GitHub: Commit Code | HTTP Request | Deployment | Quality Score OK? | Google Drive | 5. Commit, Store, Log & Notify |
|
||||
| Google Drive: Save Report | Google Drive | Documentation | GitHub: Commit Code | Google Sheets | 5. Commit, Store, Log & Notify |
|
||||
| Google Sheets: Log Result | Google Sheets | Auditing | Google Drive | Notify Slack: Success | 5. Commit, Store, Log & Notify |
|
||||
| Notify Slack: Success | Slack | Communication | Google Sheets | - | 5. Commit, Store, Log & Notify |
|
||||
| Notify Slack: Optimized | Slack | Communication | Cursor AI: Opt | - | 5. Commit, Store, Log & Notify |
|
||||
|
||||
---
|
||||
|
||||
## 4. Reproducing the Workflow from Scratch
|
||||
|
||||
1. **Global Configuration:** Create a **Set Node** ("Set Task Config"). Add fields for GitHub username, repository name, target branch (e.g., `ai-generated`), and a `quality_threshold` integer (75).
|
||||
2. **Triggers:** Add a **Webhook Node** (POST) and a **Schedule Trigger**. Connect both to the Set Node.
|
||||
3. **Classification:**
|
||||
* Connect the Set node to an **OpenAI Node**. Configure it to output a JSON summary of the task.
|
||||
* Add a **Wait Node** for 1 second.
|
||||
* Add a **Code Node** to parse the OpenAI output. Use `$input.item.json.message.content.replace(/`json|`/g, '')`.
|
||||
4. **Cursor API Integration:**
|
||||
* Create an **HTTP Request Node** ("Cursor AI: Generate Code"). URL: `https://api.cursor.sh/v1/chat/completions`. Method: POST. Use Headers for `Authorization` (Bearer Token).
|
||||
* Create a second **HTTP Request Node** for "Code Review". The body should prompt the AI to return a JSON object with a `score`.
|
||||
5. **Quality Logic:**
|
||||
* Add an **If Node**. Condition: `{{ $json.score }} >= {{ $('Set Task Config').item.json.quality_threshold }}`.
|
||||
6. **GitHub Deployment (True Path):**
|
||||
* Add an **HTTP Request Node**. Method: PUT. URL: `https://api.github.com/repos/{{owner}}/{{repo}}/contents/{{path}}`.
|
||||
* Body: Send the code as a Base64 string (use `Buffer.from().toString('base64')`).
|
||||
7. **Storage & Logging:**
|
||||
* Add a **Google Drive Node** to upload the review report.
|
||||
* Add a **Google Sheets Node** to append the row of results.
|
||||
8. **Optimization (False Path):**
|
||||
* Add an **HTTP Request Node** to Cursor AI specifically for fixing issues found in the review.
|
||||
9. **Notifications:**
|
||||
* Add **Slack Nodes** to both the "Success" and "Optimized" paths to alert the team.
|
||||
|
||||
---
|
||||
|
||||
## 5. General Notes & Resources
|
||||
|
||||
| Note Content | Context or Link |
|
||||
| :--- | :--- |
|
||||
| Cursor AI API Access | Requires Cursor developer access or a compatible proxy endpoint. |
|
||||
| GitHub Auth | Must use a Personal Access Token (PAT) with `repo` scopes. |
|
||||
| Quality Threshold | Setting this too high (95+) may trigger frequent optimization loops. |
|
||||
| Data Privacy | Ensure code sent to APIs follows your company's security policy. |
|
||||
Reference in New Issue
Block a user