klbr/n8nworkflows.xyz
1
0
Fork 0
mirror of https://github.com/khoaliber/n8nworkflows.xyz.git synced 2026-04-28 00:29:22 +00:00
Code Issues Packages Projects Releases Wiki Activity

creation

Browse Source
This commit is contained in:
nusquama
2025-11-12 20:11:45 +01:00
parent f8df379db0
commit 134fdf205b
1 changed files with 332 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
workflows/Analyze Any Website with OpenAI And Get On-Page SEO Audit-3224/readme-3224.md
+332
View File
@@ -0,0 +1,332 @@
Analyze Any Website with OpenAI And Get On-Page SEO Audit
https://n8nworkflows.xyz/workflows/analyze-any-website-with-openai-and-get-on-page-seo-audit-3224
# Analyze Any Website with OpenAI And Get On-Page SEO Audit
### 1. Workflow Overview
This workflow, titled **"Analyze Any Website with OpenAI And Get On-Page SEO Audit"**, is designed to provide an instant, comprehensive SEO audit of any landing page URL submitted by the user. It targets SaaS founders, marketing teams, agencies, e-commerce, and content sites aiming to improve their Google rankings by identifying technical and content-related SEO issues.
The workflow is logically divided into the following blocks:
- **1.1 Input Reception:** Captures the landing page URL from a user-submitted form.
- **1.2 Website Scraping:** Fetches the HTML source code of the provided URL.
- **1.3 AI-Powered SEO Audits:** Two parallel AI agents analyze the scraped HTML:
- Technical SEO Audit (HTML and on-page technical factors)
- Content SEO Audit (content quality, keywords, readability)
- **1.4 Results Aggregation and Formatting:** Merges and aggregates AI outputs, then converts them into a clean HTML report.
- **1.5 Email Notification:** Sends the audit report via Gmail to a predefined email address.
---
### 2. Block-by-Block Analysis
#### 2.1 Input Reception
- **Overview:**
This block collects the landing page URL from the user via a web form trigger, initiating the workflow.
- **Nodes Involved:**
- Landing Page Url
- **Node Details:**
- **Landing Page Url**
- Type: Form Trigger
- Role: Entry point; captures user input via a web form.
- Configuration:
- Form titled "Conversion Rate Optimizer" with a single required field labeled "Landing Page Url" (placeholder example: https://yuzuu.co).
- Form description: "Your Landing Page is Leaking Sales—Fix It Now".
- Input/Output: Outputs JSON containing the landing page URL under the key `Landing Page Url`.
- Edge Cases:
- User submits invalid or malformed URL → downstream HTTP request may fail.
- No input submitted → form enforces required field, so unlikely.
- Version: 2.2
#### 2.2 Website Scraping
- **Overview:**
Fetches the raw HTML content of the landing page URL submitted by the user.
- **Nodes Involved:**
- Scrape Website
- **Node Details:**
- **Scrape Website**
- Type: HTTP Request
- Role: Retrieves the HTML source code of the landing page.
- Configuration:
- URL dynamically set to the user input: `={{ $json['Landing Page Url'] }}`.
- Default HTTP GET method, no additional headers or options configured.
- Input: Receives URL from "Landing Page Url" node.
- Output: Returns the full HTTP response including the HTML content under `data`.
- Edge Cases:
- URL unreachable or returns error (404, 500, timeout).
- Website blocks scraping or requires authentication.
- Large page size causing timeouts or memory issues.
- Version: 4.2
#### 2.3 AI-Powered SEO Audits
- **Overview:**
Two parallel AI agents analyze the scraped HTML content: one performs a technical SEO audit, the other a content SEO audit. Both use OpenAI GPT-4o-mini model via Langchain integration.
- **Nodes Involved:**
- OpenAI Chat Model
- OpenAI Chat Model1
- Technical Audit
- Content Audit
- **Node Details:**
- **OpenAI Chat Model**
- Type: Langchain OpenAI Chat Model
- Role: Provides language model interface for the Technical Audit agent.
- Configuration:
- Model: GPT-4o-mini (optimized for cost and performance).
- Credentials: OpenAI API key configured.
- Input: Receives prompts from "Technical Audit" node.
- Output: Returns AI-generated audit results.
- Edge Cases:
- API key invalid or quota exceeded → authentication errors.
- Model timeout or rate limiting.
- Version: 1.2
- **OpenAI Chat Model1**
- Type: Langchain OpenAI Chat Model
- Role: Provides language model interface for the Content Audit agent.
- Configuration: Same as "OpenAI Chat Model".
- Input: Receives prompts from "Content Audit" node.
- Output: Returns AI-generated audit results.
- Edge Cases: Same as above.
- Version: 1.2
- **Technical Audit**
- Type: Langchain Agent
- Role: Performs detailed technical SEO audit on the HTML content.
- Configuration:
- Prompt instructs the AI to analyze the HTML code for technical SEO issues, categorizing findings into Critical Issues, Quick Wins, and Opportunities for Improvement.
- Input text includes the scraped HTML content (`{{ $json.data }}`).
- Output is a clean, bullet-pointed audit report without introductory text.
- Input: Receives HTML content from "Scrape Website" node.
- Output: Audit text sent to "Merge" node.
- Edge Cases:
- AI may misinterpret malformed HTML.
- Large HTML content might exceed token limits.
- Version: 1.7
- **Content Audit**
- Type: Langchain Agent
- Role: Performs detailed content SEO audit on the HTML content.
- Configuration:
- Prompt instructs AI to analyze content quality, keyword usage, and readability metrics.
- Input text includes the scraped HTML content (`{{ $json.data }}`).
- Output is a structured audit with analysis and recommendations in bullet points.
- Input: Receives HTML content from "Scrape Website" node.
- Output: Audit text sent to "Merge" node.
- Edge Cases:
- AI may misinterpret content if HTML is complex or heavily scripted.
- Token limits may truncate output.
- Version: 1.7
#### 2.4 Results Aggregation and Formatting
- **Overview:**
Combines the two audit reports into a single structured output and formats it as HTML for readability.
- **Nodes Involved:**
- Merge
- Aggregate
- Markdown
- **Node Details:**
- **Merge**
- Type: Merge
- Role: Combines outputs from Technical Audit and Content Audit nodes into one data stream.
- Configuration: Default merge mode (likely wait for all inputs).
- Input: Receives two inputs — Technical Audit (index 0) and Content Audit (index 1).
- Output: Passes combined data to Aggregate node.
- Edge Cases:
- One audit fails or delays → merge may timeout or output incomplete data.
- Version: 3
- **Aggregate**
- Type: Aggregate
- Role: Aggregates the merged outputs, specifically aggregating the "output" field from both audits.
- Configuration: Aggregates field named "output".
- Input: Receives merged data from Merge node.
- Output: Passes aggregated data to Markdown node.
- Edge Cases:
- Empty or missing fields could cause aggregation errors.
- Version: 1
- **Markdown**
- Type: Markdown
- Role: Converts combined markdown audit reports into HTML format for email readability.
- Configuration:
- Mode set to "markdownToHtml".
- Template includes two sections:
- "# On-Page Technical Audit" with first audit output
- "# On-Page SEO Content Audit" with second audit output
- Uses expressions to insert aggregated audit texts.
- Input: Receives aggregated audit data from Aggregate node.
- Output: HTML formatted audit report sent to Gmail node.
- Edge Cases:
- Malformed markdown could produce invalid HTML.
- Version: 1
#### 2.5 Email Notification
- **Overview:**
Sends the final SEO audit report via Gmail to a specified recipient.
- **Nodes Involved:**
- Gmail
- Sticky Note (for user instruction)
- **Node Details:**
- **Gmail**
- Type: Gmail
- Role: Sends email with the SEO audit report.
- Configuration:
- Recipient email hardcoded as "hello@youremail.com" (user must update).
- Subject line dynamically includes the landing page URL.
- Message body contains the HTML audit report from Markdown node.
- Uses OAuth2 credentials for Gmail authentication.
- Input: Receives HTML content from Markdown node.
- Output: Sends email, no further output.
- Edge Cases:
- Invalid or expired Gmail OAuth2 credentials → authentication failure.
- Email quota limits or network issues.
- Version: 2.1
- **Sticky Note**
- Type: Sticky Note
- Role: Provides user instruction to connect Gmail credentials for email sending.
- Content: "## Send Email \nConnect your credentials & Easily send emails from a Gmail address."
- Position: Near Gmail node for visual guidance.
---
### 3. Summary Table
| Node Name | Node Type | Functional Role | Input Node(s) | Output Node(s) | Sticky Note |
|--------------------|--------------------------------------|---------------------------------|-----------------------|-----------------------|----------------------------------------------------------------------------------------------|
| Landing Page Url | Form Trigger | User input of landing page URL | — | Scrape Website | |
| Scrape Website | HTTP Request | Fetches HTML source of URL | Landing Page Url | Content Audit, Technical Audit | |
| OpenAI Chat Model | Langchain OpenAI Chat Model | AI model for Technical Audit | Technical Audit (ai_languageModel) | Technical Audit | ## Open AI Setup<br>- Add your credentials<br>- Select o1 model for (way) better results.<br>- One run = one page audit = around $0.3 with o1 |
| OpenAI Chat Model1 | Langchain OpenAI Chat Model | AI model for Content Audit | Content Audit (ai_languageModel) | Content Audit | ## Open AI Setup<br>- Add your credentials<br>- Select o1 model for (way) better results.<br>- One run = one page audit = around $0.3 with o1 |
| Technical Audit | Langchain Agent | Performs technical SEO audit | Scrape Website, OpenAI Chat Model | Merge | ## Open AI Setup<br>- Add your credentials<br>- Select o1 model for (way) better results.<br>- One run = one page audit = around $0.3 with o1 |
| Content Audit | Langchain Agent | Performs content SEO audit | Scrape Website, OpenAI Chat Model1 | Merge | ## Open AI Setup<br>- Add your credentials<br>- Select o1 model for (way) better results.<br>- One run = one page audit = around $0.3 with o1 |
| Merge | Merge | Combines audit outputs | Technical Audit, Content Audit | Aggregate | |
| Aggregate | Aggregate | Aggregates audit outputs | Merge | Markdown | |
| Markdown | Markdown | Converts markdown to HTML | Aggregate | Gmail | |
| Gmail | Gmail | Sends audit report via email | Markdown | — | ## Send Email<br>Connect your credentials & Easily send emails from a Gmail address. |
| Sticky Note | Sticky Note | Instruction for Gmail setup | — | — | ## Send Email<br>Connect your credentials & Easily send emails from a Gmail address. |
| Sticky Note1 | Sticky Note | Instruction for OpenAI setup | — | — | ## Open AI Setup<br>- Add your credentials<br>- Select o1 model for (way) better results.<br>- One run = one page audit = around $0.3 with o1 |
---
### 4. Reproducing the Workflow from Scratch
1. **Create a Form Trigger Node**
- Name: "Landing Page Url"
- Type: Form Trigger
- Configure form title: "Conversion Rate Optimizer"
- Add one required form field:
- Label: "Landing Page Url"
- Placeholder: "https://yuzuu.co"
- Add form description: "Your Landing Page is Leaking Sales—Fix It Now"
- This node will start the workflow when a user submits a URL.
2. **Add an HTTP Request Node**
- Name: "Scrape Website"
- Type: HTTP Request
- Set URL parameter to expression: `={{ $json["Landing Page Url"] }}`
- Method: GET (default)
- No authentication or headers needed unless target site requires it.
- Connect output of "Landing Page Url" to input of "Scrape Website".
3. **Add Two Langchain OpenAI Chat Model Nodes**
- Names: "OpenAI Chat Model" and "OpenAI Chat Model1"
- Type: Langchain OpenAI Chat Model
- Model: Select "gpt-4o-mini" or equivalent OpenAI model optimized for cost/performance.
- Credentials: Configure OpenAI API key credentials for both nodes.
- These nodes serve as language model interfaces for the two AI agents.
4. **Add Two Langchain Agent Nodes**
- Names: "Technical Audit" and "Content Audit"
- Type: Langchain Agent
- For "Technical Audit":
- Set prompt text to instruct AI to analyze HTML for technical SEO issues, categorizing findings into Critical Issues, Quick Wins, and Opportunities for Improvement.
- Use expression to pass scraped HTML content: `{{ $json.data }}`
- Set prompt type to "define".
- Connect "OpenAI Chat Model" node to "Technical Audit" node's AI language model input.
- For "Content Audit":
- Set prompt text to instruct AI to analyze content quality, keyword usage, and readability metrics.
- Use expression to pass scraped HTML content: `{{ $json.data }}`
- Set prompt type to "define".
- Connect "OpenAI Chat Model1" node to "Content Audit" node's AI language model input.
- Connect output of "Scrape Website" node to both "Technical Audit" and "Content Audit" nodes.
5. **Add a Merge Node**
- Name: "Merge"
- Type: Merge
- Default merge mode (wait for all inputs)
- Connect outputs of "Technical Audit" (index 0) and "Content Audit" (index 1) to inputs of "Merge".
6. **Add an Aggregate Node**
- Name: "Aggregate"
- Type: Aggregate
- Configure to aggregate the field named "output" from incoming data.
- Connect output of "Merge" to input of "Aggregate".
7. **Add a Markdown Node**
- Name: "Markdown"
- Type: Markdown
- Mode: markdownToHtml
- Markdown template:
```
# On-Page Technical Audit
{{ $json.output[0] }}
# On-Page SEO Content Audit
{{ $json.output[1] }}
```
- Connect output of "Aggregate" to input of "Markdown".
8. **Add a Gmail Node**
- Name: "Gmail"
- Type: Gmail
- Configure OAuth2 credentials for Gmail account.
- Set recipient email (e.g., "hello@youremail.com") — replace with your actual email.
- Subject: Use expression to include landing page URL, e.g., `=On-Page SEO Audit - {{ $('Landing Page Url').item.json['Landing Page Url'] }}`
- Message: Use expression to pass HTML content from "Markdown" node, e.g., `={{ $json.data }}`
- Connect output of "Markdown" to input of "Gmail".
9. **Add Sticky Notes for User Guidance** (optional but recommended)
- Near OpenAI nodes: Add note explaining to add OpenAI credentials and use the GPT-4o-mini or o1 model for best results and cost awareness.
- Near Gmail node: Add note reminding to connect Gmail credentials for sending emails.
10. **Set Execution Order**
- Ensure the workflow executes sequentially from form trigger → HTTP request → AI audits → merge → aggregate → markdown → email.
---
### 5. General Notes & Resources
| Note Content | Context or Link |
|----------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------|
| The workflow uses OpenAI GPT-4o-mini (o1) model for balanced cost and performance in SEO audits. | OpenAI API documentation and pricing details. |
| One run of the workflow (one page audit) costs approximately $0.20–$0.30 with the selected model. | Budget planning for API usage. |
| Update the Gmail node recipient email to your actual email address before running the workflow. | Prevents sending audit reports to placeholder addresses. |
| The workflow requires valid OpenAI API credentials and Gmail OAuth2 credentials configured in n8n. | Credential setup instructions in n8n documentation. |
| Example audit output screenshot is included in the original workflow description (not embedded here). | Visual reference for expected audit report format. |
| SEO audits cover both technical on-page factors and content quality/readability for comprehensive analysis. | Best practice for holistic SEO improvements. |
---
This documentation provides a detailed, structured reference for understanding, reproducing, and maintaining the "Analyze Any Website with OpenAI And Get On-Page SEO Audit" workflow. It anticipates common failure points such as invalid URLs, API errors, and credential misconfigurations, enabling robust operation and easy customization.