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
2025-11-13 13:18:48 +01:00
parent 1b78910841
commit ea5661370d
1 changed files with 423 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
workflows/Summarize Content from URLs, Text & PDFs using OpenAI GPT-4.1-mini-8210/readme-8210.md
+423
View File
@@ -0,0 +1,423 @@
Summarize Content from URLs, Text & PDFs using OpenAI GPT-4.1-mini
https://n8nworkflows.xyz/workflows/summarize-content-from-urls--text---pdfs-using-openai-gpt-4-1-mini-8210
# Summarize Content from URLs, Text & PDFs using OpenAI GPT-4.1-mini
### 1. Workflow Overview
This workflow is an AI-powered content summarization suite designed to handle three distinct input types: URLs, raw text, and PDF files. Its primary purpose is to extract, process, and summarize content from these sources using OpenAI’s GPT-4.1-mini model, producing concise, customizable summaries in markdown format. The workflow supports multi-language summaries, summary length options (brief, standard, detailed), and focus areas such as key points, numbers/data, conclusions, or action items.
The workflow is logically divided into four main blocks:
- **1.1 Input Reception & Routing:** Receives incoming requests via webhooks, parses input parameters, and directs processing based on input type (URL, text, or file).
- **1.2 URL Content Processing:** Fetches and cleans content from web URLs, then summarizes it.
- **1.3 Text Content Processing:** Processes raw text inputs directly for summarization.
- **1.4 PDF File Processing:** Downloads PDFs, extracts text (with fallback OCR), and summarizes the extracted content.
Each processing block culminates in a formatted summary response sent back via webhook.
---
### 2. Block-by-Block Analysis
#### 1.1 Input Reception & Routing
**Overview:**
This block receives incoming HTTP requests, extracts and normalizes input parameters, and routes the workflow based on whether the input is a URL, raw text, or a PDF file.
**Nodes Involved:**
- Initial Trigger
- Parse Input Parameters
- Input Type Switch
- Clean & Format Content (used for text and partially for URL)
**Node Details:**
- **Initial Trigger**
- Type: Webhook
- Role: Entry point webhook triggering the workflow on HTTP requests.
- Config: Receives raw body, responds via response node.
- Inputs: None (trigger)
- Outputs: Parse Input Parameters
- Edge cases: Invalid HTTP requests, missing body or query params.
- **Parse Input Parameters**
- Type: Set
- Role: Extracts and sets normalized input parameters (input_type, url, text_content, summary_length, focus, language, file_content) from query or request body.
- Config: Uses expressions to handle missing values and provide defaults (e.g., input_type defaults to "url", summary_length to "standard").
- Inputs: Initial Trigger
- Outputs: Input Type Switch
- Edge cases: Missing or malformed parameters.
- **Input Type Switch**
- Type: Switch
- Role: Routes execution based on input_type value ("url", "text", or "file").
- Config: Strict string equality checks on `$json.input_type`.
- Inputs: Parse Input Parameters
- Outputs:
- "url" → Fetch URL Content
- "text" → Clean & Format Content
- "file" → Fetch PDF File
- Edge cases: Unknown input_type values, case sensitivity.
- **Clean & Format Content**
- Type: Set
- Role: Normalizes and prepares raw content for AI summarization by setting a unified "raw_content" field from multiple possible sources.
- Inputs: From Input Type Switch (for "text") and OCR.Space node (fallback)
- Outputs: Generate AI Summary (Unified)
- Edge cases: Empty or malformed content fields.
---
#### 1.2 URL Content Processing
**Overview:**
Handles input URLs by scraping webpage content, cleaning it, and generating a summary with AI.
**Nodes Involved:**
- summarize-url (Webhook)
- Extract URL Parameters
- Scrape Web Page Content
- Clean Scraped Content
- Summarize URL Content (Langchain Agent)
- OpenAI GPT-4.1 (URL)
- Format URL Summary
- Return URL Summary
**Node Details:**
- **summarize-url**
- Type: Webhook
- Role: Entry point for URL summarization requests, secured with header authentication.
- Config: Raw body, authentication via HTTP header bearer token.
- Outputs: Extract URL Parameters
- Edge cases: Auth failures, invalid webhook requests.
- **Extract URL Parameters**
- Type: Set
- Role: Extracts request parameters for URL summarization with defaults.
- Inputs: summarize-url
- Outputs: Scrape Web Page Content
- **Scrape Web Page Content**
- Type: HTTP Request
- Role: Fetches raw HTML/text content from the given URL.
- Config: URL from `$json.url`.
- Inputs: Extract URL Parameters
- Outputs: Clean Scraped Content
- Edge cases: HTTP errors, 404, redirects, timeouts.
- **Clean Scraped Content**
- Type: Set
- Role: Normalizes fetched webpage content into a "raw_content" string for summarization.
- Inputs: Scrape Web Page Content
- Outputs: Summarize URL Content
- **Summarize URL Content**
- Type: Langchain Agent
- Role: Defines an AI prompt to summarize content with dynamic length, focus, and language based on parameters.
- Config: System message includes instructions with markdown formatting and language selection.
- Inputs: Clean Scraped Content
- Outputs: OpenAI GPT-4.1 (URL)
- **OpenAI GPT-4.1 (URL)**
- Type: Langchain LM Chat (OpenAI)
- Role: Calls OpenAI GPT-4.1-mini model to generate the summary text.
- Config: Uses stored OpenAI credentials.
- Inputs: Summarize URL Content
- Outputs: Format URL Summary
- Edge cases: API rate limits, auth errors, invalid prompts.
- **Format URL Summary**
- Type: Set
- Role: Sets the AI output into a "content" field for response.
- Inputs: OpenAI GPT-4.1 (URL)
- Outputs: Return URL Summary
- **Return URL Summary**
- Type: Respond to Webhook
- Role: Sends back the final summary text as HTTP response.
- Inputs: Format URL Summary
- Outputs: None
---
#### 1.3 Text Content Processing
**Overview:**
Processes raw text input directly to produce AI-generated summaries.
**Nodes Involved:**
- summarize-text (Webhook)
- Extract Text Parameters
- Process Raw Text Input
- Summarize Text Input (Langchain Agent)
- OpenAI GPT-4.1 (Text)
- Format Text Summary
- Return Text Summary
**Node Details:**
- **summarize-text**
- Type: Webhook
- Role: Entry point for raw text summarization requests with header authentication.
- Inputs: HTTP request
- Outputs: Extract Text Parameters
- **Extract Text Parameters**
- Type: Set
- Role: Normalizes input parameters for text summarization.
- Outputs: Process Raw Text Input
- **Process Raw Text Input**
- Type: Set
- Role: Sets "raw_content" from varied possible fields containing the text.
- Outputs: Summarize Text Input
- **Summarize Text Input**
- Type: Langchain Agent
- Role: Prepares AI prompt with summarization instructions similar to URL summarization but for raw text.
- Outputs: OpenAI GPT-4.1 (Text)
- **OpenAI GPT-4.1 (Text)**
- Type: Langchain LM Chat (OpenAI)
- Role: Calls GPT-4.1-mini for text summarization.
- Outputs: Format Text Summary
- **Format Text Summary**
- Type: Set
- Role: Maps AI output to "content".
- Outputs: Return Text Summary
- **Return Text Summary**
- Type: Respond to Webhook
- Role: Returns summary via HTTP response.
---
#### 1.4 PDF File Processing
**Overview:**
Downloads PDF files, attempts text extraction, falls back to OCR if needed, and summarizes the extracted text.
**Nodes Involved:**
- summarize-file (Webhook)
- Extract File Parameters
- Download PDF File
- Extract Text from PDF
- Check PDF Text Extraction (If)
- Map OCR Language Code
- OCR Fallback Processing (HTTP Request to OCR.Space)
- Process Extracted Text
- Summarize PDF Content (Langchain Agent)
- OpenAI GPT-4.1 (PDF)
- Format PDF Summary
- Return PDF Summary
**Node Details:**
- **summarize-file**
- Type: Webhook
- Role: Entry point for PDF summarization requests, header authenticated.
- Outputs: Extract File Parameters
- **Extract File Parameters**
- Type: Set
- Role: Normalizes input including file URL and summarization options.
- Outputs: Download PDF File
- **Download PDF File**
- Type: HTTP Request
- Role: Downloads the PDF binary from the provided URL.
- Outputs: Extract Text from PDF
- Edge cases: URL inaccessible, network errors.
- **Extract Text from PDF**
- Type: Extract From File
- Role: Extracts text content from PDF binary.
- Outputs: Check PDF Text Extraction
- **Check PDF Text Extraction**
- Type: If
- Role: Checks if extracted text is empty.
- If True (empty): Map OCR Language Code → OCR Fallback Processing
- If False (text found): Process Extracted Text
- **Map OCR Language Code**
- Type: Function
- Role: Converts language name to OCR.Space language code (e.g., 'english' → 'eng').
- Outputs: OCR Fallback Processing
- **OCR Fallback Processing**
- Type: HTTP Request
- Role: Sends PDF to OCR.Space API for OCR-based text extraction.
- Requires OCR.Space API key configured in header.
- Outputs: Process Extracted Text
- Edge cases: OCR API limit, invalid file format.
- **Process Extracted Text**
- Type: Set
- Role: Normalizes OCR response or extracted text to "raw_content".
- Outputs: Summarize PDF Content
- **Summarize PDF Content**
- Type: Langchain Agent
- Role: Prepares AI prompt to summarize PDF text with summary length, focus, and language options.
- Outputs: OpenAI GPT-4.1 (PDF)
- **OpenAI GPT-4.1 (PDF)**
- Type: Langchain LM Chat (OpenAI)
- Role: Calls GPT-4.1-mini for PDF content summarization.
- Outputs: Format PDF Summary
- **Format PDF Summary**
- Type: Set
- Role: Maps AI output to "content".
- Outputs: Return PDF Summary
- **Return PDF Summary**
- Type: Respond to Webhook
- Role: Returns final summary response via HTTP.
- **Sticky Note4:** Important: PDF must be publicly accessible for download.
---
#### Additional Unified Summarization (Not explicitly triggered in this export but present)
- **Clean & Format Content** node normalizes content for a unified summarization path.
- **Generate AI Summary (Unified)** and **OpenAI GPT-4.1 (Unified)** nodes handle AI summarization for this unified path.
- **Structure Final Output** and **Return Summary Response** nodes finalize and respond with the summary.
---
### 3. Summary Table
| Node Name | Node Type | Functional Role | Input Node(s) | Output Node(s) | Sticky Note |
|----------------------------|---------------------------------|-----------------------------------------------|----------------------------------|---------------------------------|-----------------------------------------------------------|
| Initial Trigger | Webhook | Entry point webhook for all inputs | None | Parse Input Parameters | # Summarize All in One |
| Parse Input Parameters | Set | Extracts and normalizes input parameters | Initial Trigger | Input Type Switch | # Summarize All in One |
| Input Type Switch | Switch | Routes based on input_type | Parse Input Parameters | Fetch URL Content, Clean & Format Content, Fetch PDF File | # Summarize All in One |
| Fetch URL Content | HTTP Request | Fetches raw content from URL | Input Type Switch (url) | Clean & Format Content | |
| Clean & Format Content | Set | Normalizes raw content | Fetch URL Content, OCR.Space | Generate AI Summary (Unified) | |
| Generate AI Summary (Unified)| Langchain Agent | Prepares AI prompt for unified summarization | Clean & Format Content | OpenAI GPT-4.1 (Unified) | |
| OpenAI GPT-4.1 (Unified) | Langchain LM Chat (OpenAI) | Calls GPT-4.1-mini for AI summary | Generate AI Summary (Unified) | Structure Final Output | |
| Structure Final Output | Set | Prepares final content field | Generate AI Summary (Unified) | Return Summary Response | |
| Return Summary Response | Respond to Webhook | Sends final summary response | Structure Final Output | None | |
| summarize-url | Webhook | Entry point for URL summarization | None | Extract URL Parameters | # Summarize URL |
| Extract URL Parameters | Set | Extracts and normalizes URL parameters | summarize-url | Scrape Web Page Content | # Summarize URL |
| Scrape Web Page Content | HTTP Request | Fetches webpage content | Extract URL Parameters | Clean Scraped Content | # Summarize URL |
| Clean Scraped Content | Set | Normalizes scraped content | Scrape Web Page Content | Summarize URL Content | # Summarize URL |
| Summarize URL Content | Langchain Agent | Prepares AI prompt for URL content | Clean Scraped Content | OpenAI GPT-4.1 (URL) | # Summarize URL |
| OpenAI GPT-4.1 (URL) | Langchain LM Chat (OpenAI) | Calls GPT-4.1-mini for URL summary | Summarize URL Content | Format URL Summary | # Summarize URL |
| Format URL Summary | Set | Sets AI output to content field | OpenAI GPT-4.1 (URL) | Return URL Summary | # Summarize URL |
| Return URL Summary | Respond to Webhook | Sends URL summary response | Format URL Summary | None | # Summarize URL |
| summarize-text | Webhook | Entry point for text summarization | None | Extract Text Parameters | # Summarize text |
| Extract Text Parameters | Set | Extracts and normalizes text parameters | summarize-text | Process Raw Text Input | # Summarize text |
| Process Raw Text Input | Set | Sets raw_content for summarization | Extract Text Parameters | Summarize Text Input | # Summarize text |
| Summarize Text Input | Langchain Agent | Prepares AI prompt for text summarization | Process Raw Text Input | OpenAI GPT-4.1 (Text) | # Summarize text |
| OpenAI GPT-4.1 (Text) | Langchain LM Chat (OpenAI) | Calls GPT-4.1-mini for text summary | Summarize Text Input | Format Text Summary | # Summarize text |
| Format Text Summary | Set | Sets AI output to content field | OpenAI GPT-4.1 (Text) | Return Text Summary | # Summarize text |
| Return Text Summary | Respond to Webhook | Sends text summary response | Format Text Summary | None | # Summarize text |
| summarize-file | Webhook | Entry point for PDF summarization | None | Extract File Parameters | # Summarize a PDF file |
| Extract File Parameters | Set | Extracts and normalizes file parameters | summarize-file | Download PDF File | # Summarize a PDF file |
| Download PDF File | HTTP Request | Downloads PDF binary from URL | Extract File Parameters | Extract Text from PDF | # Summarize a PDF file |
| Extract Text from PDF | Extract From File | Extracts text from PDF binary | Download PDF File | Check PDF Text Extraction | # Summarize a PDF file |
| Check PDF Text Extraction | If | Checks if extracted text is empty | Extract Text from PDF | Map OCR Language Code / Process Extracted Text | # Summarize a PDF file |
| Map OCR Language Code | Function | Converts language to OCR.Space code | Check PDF Text Extraction (True) | OCR Fallback Processing | # Summarize a PDF file |
| OCR Fallback Processing | HTTP Request | Calls OCR.Space API for OCR extraction | Map OCR Language Code | Process Extracted Text | # Summarize a PDF file |
| Process Extracted Text | Set | Normalizes OCR or extracted text | OCR Fallback Processing / Check PDF Text Extraction (False) | Summarize PDF Content | # Summarize a PDF file |
| Summarize PDF Content | Langchain Agent | Prepares AI prompt for PDF content summarization | Process Extracted Text | OpenAI GPT-4.1 (PDF) | # Summarize a PDF file |
| OpenAI GPT-4.1 (PDF) | Langchain LM Chat (OpenAI) | Calls GPT-4.1-mini for PDF summary | Summarize PDF Content | Format PDF Summary | # Summarize a PDF file |
| Format PDF Summary | Set | Sets AI output to content field | OpenAI GPT-4.1 (PDF) | Return PDF Summary | # Summarize a PDF file |
| Return PDF Summary | Respond to Webhook | Sends PDF summary response | Format PDF Summary | None | # Summarize a PDF file |
| Map Language Code | Function | Maps language input to 3-letter code for OCR | Check If Extracted Text Empty | OCR.Space | |
| OCR.Space | HTTP Request | Calls OCR.Space API for image-based text extraction | Map Language Code | Clean & Format Content | |
---
### 4. Reproducing the Workflow from Scratch
1. **Create Webhook Nodes for Input Types:**
- `Initial Trigger`: Webhook to receive all inputs (path: "webhook-url", raw body enabled, response mode: response node).
- `summarize-url`: Webhook for URL summarization (secure with header bearer auth).
- `summarize-text`: Webhook for text summarization (secure with header bearer auth).
- `summarize-file`: Webhook for PDF summarization (secure with header bearer auth).
2. **Parse Input Parameters:**
- Create three Set nodes (`Parse Input Parameters`, `Extract URL Parameters`, `Extract Text Parameters`, `Extract File Parameters`) that extract and normalize inputs:
- Parameters: input_type (default "url"), url, text_content, summary_length (default "standard"), focus (default "key_points"), language (default "english"), file_content.
- Connect each webhook to its respective parameter extraction node.
3. **Input Routing:**
- From `Parse Input Parameters`, create a Switch node `Input Type Switch` that routes based on `input_type`:
- "url" → `Fetch URL Content`
- "text" → `Clean & Format Content`
- "file" → `Fetch PDF File`
4. **URL Processing:**
- `Fetch URL Content`: HTTP Request node fetching content from `$json.url`.
- `Clean Scraped Content`: Set node to set `raw_content` from HTTP response.
- `Summarize URL Content`: Langchain Agent node with prompt instructions referencing `summary_length`, `focus`, `language`.
- `OpenAI GPT-4.1 (URL)`: Langchain LM Chat node calling GPT-4.1-mini with OpenAI credentials.
- `Format URL Summary`: Set node mapping AI output to `content`.
- `Return URL Summary`: Respond to Webhook node returning `content`.
- Connect nodes in listed order starting from `Extract URL Parameters`.
5. **Text Processing:**
- `Process Raw Text Input`: Set node to assign `raw_content` from input text.
- `Summarize Text Input`: Langchain Agent node with similar prompt logic.
- `OpenAI GPT-4.1 (Text)`: LM Chat node calling GPT-4.1-mini.
- `Format Text Summary`: Set node for output mapping.
- `Return Text Summary`: Respond to Webhook node.
- Connect nodes from `Extract Text Parameters`.
6. **PDF Processing:**
- `Download PDF File`: HTTP Request node downloading PDF from URL.
- `Extract Text from PDF`: Extract From File node set to PDF operation.
- `Check PDF Text Extraction`: If node checking if extracted text is empty.
- On True branch:
- `Map OCR Language Code`: Function node converting language to OCR code (eng, spa, fra, ger).
- `OCR Fallback Processing`: HTTP Request node calling OCR.Space API with API key in headers, posting multipart-form-data with file and language.
- `Process Extracted Text`: Set node setting `raw_content` from OCR response.
- On False branch:
- `Process Extracted Text`: Set node setting `raw_content` from extracted text.
- `Summarize PDF Content`: Langchain Agent node with prompt logic for PDF summarization.
- `OpenAI GPT-4.1 (PDF)`: LM Chat node with GPT-4.1-mini.
- `Format PDF Summary`: Set node mapping AI output.
- `Return PDF Summary`: Respond to Webhook node.
- Connect nodes from `Extract File Parameters`.
7. **Unified Summarization (Optional):**
- Connect `Clean & Format Content` to `Generate AI Summary (Unified)` Langchain Agent node.
- Connect to `OpenAI GPT-4.1 (Unified)`, then `Structure Final Output` and `Return Summary Response`.
- Use this path for a combined all-in-one summarization, useful if input type detection and routing is done upstream.
8. **Credential Setup:**
- Configure OpenAI API credentials and assign to all OpenAI nodes.
- Configure OCR.Space API key in the OCR HTTP Request node headers.
- Configure HTTP Header Authentication credentials for webhook security.
9. **General Configuration:**
- Use appropriate webhook URLs and authentication for your environment.
- Ensure PDF URLs are publicly accessible.
- Adjust prompt templates for custom summarization needs.
- Test each path independently.
---
### 5. General Notes & Resources
| Note Content | Context or Link |
|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------|
| AI Content Summarizer Suite: This n8n template collection demonstrates a comprehensive AI-powered content summarization system handling URLs, text, and PDFs with fallback OCR. Use cases include research, content curation, document processing, meeting prep, and social media content creation. Customizable summary length and focus with multi-language support. Integration-ready for platforms like Bubble and Zapier. Requirements: OpenAI API key, OCR.Space API key (optional for PDFs), and n8n instance. Setup involves credential replacement and webhook configuration. | Provided in Sticky Note6 content inside workflow. |
| Important: PDF files must be publicly accessible URLs for download and processing. | From Sticky Note4 near PDF processing nodes. |
| Supports multi-language (English, Spanish, French, German) with dynamic prompt adaptation for summary style and focus areas (key points, numbers, conclusions, action items). Formatting guidelines enforce markdown with bold, italics, bullet points, and section headers for clarity. | From prompt templates in Langchain Agent nodes. |
| OAuth2 or other authentication for webhooks can be configured or disabled depending on internal/external use. | From webhook node configurations and Sticky Note6 instructions. |
| This workflow is designed modularly with four separate webhooks allowing flexible integration and testing per input type, plus an all-in-one entry point for unified summarization. | Workflow design overview. |
---
**Disclaimer:** The text provided is exclusively generated from an automated workflow created with n8n integration and automation tool. It strictly complies with content policies and contains no illegal, offensive, or protected elements. All data processed is legal and public.