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
2025-12-13 05:01:12 +01:00
parent ddbd7b2e32
commit eb91c1b8ed
1 changed files with 323 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
workflows/Create AI Diary Entries from LINE Photos with OpenAI Vision and Google Drive-11445/readme-11445.md
+323
View File
@@ -0,0 +1,323 @@
Create AI Diary Entries from LINE Photos with OpenAI Vision and Google Drive
https://n8nworkflows.xyz/workflows/create-ai-diary-entries-from-line-photos-with-openai-vision-and-google-drive-11445
# Create AI Diary Entries from LINE Photos with OpenAI Vision and Google Drive
### 1. Workflow Overview
This workflow automates the creation of AI-generated diary entries from photos sent via the LINE messaging app. When a user sends a photo to a LINE bot, the workflow:
- Receives the photo via a webhook
- Extracts necessary metadata from the LINE message event
- Downloads the photo content from LINE
- Uses OpenAI Vision capabilities to generate a short Japanese diary-style sentence describing the photo
- Prepares a dated folder structure on Google Drive (`KidsDiary/YYYY/MM`)
- Uploads both the original photo and a text file containing the AI-generated diary sentence into the corresponding Google Drive folder
- Sends a confirmation reply back to the user on LINE
**Target Use Cases:**
- Parents or caregivers wanting an automatic kids’ photo diary
- Users who want to archive LINE photos with contextual AI-generated notes
- Anyone preferring a “take a photo and forget” journaling style
---
The workflow logic is grouped into four main blocks:
- **1.1 Input Reception:** Receiving and extracting photo data from LINE
- **1.2 AI Diary Generation & Path Preparation:** Using OpenAI Vision to generate diary text and preparing Google Drive folder paths
- **1.3 Diary File Preparation:** Combining data and creating a text file for Google Drive
- **1.4 Storage & Confirmation:** Uploading files to Google Drive and replying to the user on LINE
---
### 2. Block-by-Block Analysis
#### 1.1 Input Reception
**Overview:**
Handles receiving the photo message from LINE, extracting relevant metadata, and downloading the photo binary.
**Nodes Involved:**
- LINE Webhook
- Extract LINE Message Data
- Get Image from LINE
**Node Details:**
- **LINE Webhook**
- Type: Webhook (n8n native)
- Role: Entry point receiving POST requests from LINE's messaging API when users send messages
- Configuration: Path set to `line-webhook`, HTTP method POST, response from last node
- Input: External HTTP POST from LINE platform
- Output: JSON event payload containing message data
- Edge Cases: Missing events, invalid webhook setup, unauthorized requests if LINE signature check not implemented (not shown here)
- **Extract LINE Message Data**
- Type: Code (JavaScript)
- Role: Parse the incoming webhook JSON to extract `messageId`, `replyToken`, and message type for further processing
- Key Expressions: Accesses `$input.first().json.body.events` array, safely returns empty if no events
- Input: JSON event body from LINE Webhook
- Output: JSON with extracted fields and full event for reference
- Edge Cases: Empty events array, missing message or fields, malformed JSON
- **Get Image from LINE**
- Type: HTTP Request
- Role: Calls LINE Content API to download the photo content in binary file format
- Configuration: URL templated with `messageId`, Authorization header with bearer token (placeholder `YOUR_TOKEN_HERE` to be replaced)
- Input: Extracted `messageId` from previous node
- Output: Binary photo data
- Edge Cases: Authorization failure, invalid/expired token, missing or invalid messageId, network timeout
---
#### 1.2 AI Diary Generation & Path Preparation
**Overview:**
Uses OpenAI Vision to generate a concise diary sentence based on the photo and prepares the folder path for storage on Google Drive.
**Nodes Involved:**
- OpenAI Vision - Generate Diary
- Prepare Folder Path
- Create Folder Structure
**Node Details:**
- **OpenAI Vision - Generate Diary**
- Type: OpenAI node via LangChain integration
- Role: Analyzes the photo image (input as base64 binary) and generates a short diary sentence in Japanese (~50 characters)
- Configuration: Model set to `gpt-4o`, operation `analyze` on image input, prompt text instructs diary style output
- Input: Binary photo content from "Get Image from LINE"
- Output: JSON with AI-generated diary sentence in `content` property
- Edge Cases: API quota limits, latency/timeouts, malformed image data, unexpected output formatting
- **Prepare Folder Path**
- Type: Code (JavaScript)
- Role: Builds the Google Drive folder path string based on current date (`KidsDiary/YYYY/MM`), extracts diary text from AI output, and bundles binary image data for next steps
- Key Expressions: Uses `Date` object, pads month, accesses previous node output for diary text and binary image
- Input: AI-generated diary text and image binary
- Output: JSON with folder path details and binary image data attached for file uploads
- Edge Cases: Missing diary text fallback to empty string, date/time inconsistencies
- **Create Folder Structure**
- Type: Google Drive node
- Role: Creates (or ensures existence of) the folder on Google Drive for the current year/month under `KidsDiary`
- Configuration: Folder name dynamically built as `KidsDiary_YYYY_MM`, created at root drive folder
- Input: Folder path info from "Prepare Folder Path"
- Output: Folder metadata including folder ID
- Edge Cases: Google Drive API rate limits, insufficient permissions, folder name conflicts
---
#### 1.3 Diary File Preparation
**Overview:**
Combines AI diary text and image metadata, converts diary text to a `.txt` file, and prepares it for upload.
**Nodes Involved:**
- Merge
- Convert to File
- Upload Photo to diary　new
**Node Details:**
- **Merge**
- Type: Merge (Combine mode)
- Role: Joins outputs from "Create Folder Structure" (folder info) and from "Prepare Folder Path" (binary image and diary text) into one data set for file upload
- Input: Two inputs, combined by position
- Output: Merged JSON and binary data
- Edge Cases: Misaligned inputs, missing data in either input
- **Convert to File**
- Type: Convert To File
- Role: Converts diary text string into a text file (binary) for Google Drive upload
- Configuration: Operation set to `toText`, source property is `diaryText` from merged data
- Input: Merged diary text JSON
- Output: Binary text file data
- Edge Cases: Empty diary text producing empty files
- **Upload Photo to diary　new**
- Type: Google Drive node
- Role: Uploads the diary text `.txt` file to Google Drive, placing it inside the folder created earlier
- Configuration: File name timestamped with current date/time followed by `_diary.txt`, folder ID set dynamically from folder creation node output
- Input: Binary text file and folder ID
- Output: File metadata on Google Drive
- Edge Cases: Upload failure, permission issues, naming conflicts
---
#### 1.4 Storage & Confirmation
**Overview:**
Uploads the original photo to Google Drive and sends a confirmation message back to the LINE user.
**Nodes Involved:**
- Upload Photo to Drive
- Reply to LINE
**Node Details:**
- **Upload Photo to Drive**
- Type: Google Drive node
- Role: Uploads the original photo binary to Google Drive under the `KidsDiary/YYYY/MM` folder with AI-generated diary text as filename
- Configuration: File name set dynamically using diary text, folder ID set to root (appears inconsistent with folder creation, see notes)
- Input: Photo binary and diary text from merged data
- Output: File metadata on Google Drive
- Edge Cases: File naming conflicts, folder ID misconfiguration, upload errors
- **Reply to LINE**
- Type: HTTP Request
- Role: Sends a reply message to the LINE user confirming successful photo and diary saving
- Configuration: POST to LINE Messaging API's reply endpoint, JSON body includes `replyToken` and a fixed text message, Authorization bearer token (placeholder)
- Input: Reply token extracted from LINE message data earlier
- Output: HTTP response from LINE API
- Edge Cases: Invalid or expired reply token, authorization failures, network issues
---
### 3. Summary Table
| Node Name | Node Type | Functional Role | Input Node(s) | Output Node(s) | Sticky Note |
|-------------------------------|---------------------------|----------------------------------------------|------------------------------|------------------------------|---------------------------------------------------------------------------------------------------|
| LINE Webhook | Webhook | Receives photo messages from LINE users | External HTTP POST | Extract LINE Message Data | Step 1 – Receive photo: Handles incoming photos from LINE via webhook |
| Extract LINE Message Data | Code | Extracts messageId, replyToken, and type | LINE Webhook | Get Image from LINE | Step 1 – Receive photo: Parses event payload |
| Get Image from LINE | HTTP Request | Downloads photo binary from LINE content API | Extract LINE Message Data | OpenAI Vision - Generate Diary | Step 1 – Receive photo: Downloads image binary |
| OpenAI Vision - Generate Diary| OpenAI LangChain Node | Generates short diary text from photo | Get Image from LINE | Prepare Folder Path | Step 2 – Create diary text & folder path: Uses AI to generate diary sentence |
| Prepare Folder Path | Code | Builds folder path & prepares data for upload| OpenAI Vision - Generate Diary | Create Folder Structure, Merge | Step 2 – Create diary text & folder path: Prepares Google Drive path and bundles data |
| Create Folder Structure | Google Drive Folder | Creates folder in Google Drive | Prepare Folder Path | Merge | Step 2 – Create diary text & folder path: Ensures folder exists |
| Merge | Merge | Combines folder info and file data | Create Folder Structure, Prepare Folder Path | Convert to File, Upload Photo to Drive | Step 3 – Prepare diary file for Google Drive: Joins data for file uploads |
| Convert to File | Convert To File | Converts diary text to binary text file | Merge | Upload Photo to diary　new | Step 3 – Prepare diary file for Google Drive: Creates diary text file |
| Upload Photo to diary　new | Google Drive File Upload | Uploads diary text file to Google Drive | Convert to File | Reply to LINE | Step 3 – Prepare diary file for Google Drive: Saves diary text file |
| Upload Photo to Drive | Google Drive File Upload | Uploads original photo to Google Drive | Merge | Reply to LINE | Step 4 – Save original photo & confirm: Saves photo file |
| Reply to LINE | HTTP Request | Sends confirmation message back to LINE user | Upload Photo to diary　new, Upload Photo to Drive | None | Step 4 – Save original photo & confirm: Confirms successful save |
| Sticky Note - Step 1 | Sticky Note | Documentation for Step 1 | None | None | Step 1 – Receive photo group explanation |
| Sticky Note - Step 2 | Sticky Note | Documentation for Step 2 | None | None | Step 2 – Create diary text & folder path explanation |
| Sticky Note - Step 3 | Sticky Note | Documentation for Step 3 | None | None | Step 3 – Prepare diary file for Google Drive explanation |
| Sticky Note - Step 4 | Sticky Note | Documentation for Step 4 | None | None | Step 4 – Save original photo & confirm explanation |
| Sticky Note - Overview1 | Sticky Note | Workflow overview and user context | None | None | Overview of full workflow purpose and audience |
---
### 4. Reproducing the Workflow from Scratch
1. **Create LINE Webhook node**
- Type: Webhook
- HTTP Method: POST
- Path: `line-webhook`
- Response Mode: Last Node
- No credentials needed here, but LINE webhook URL must be registered in LINE Developer Console
2. **Create Extract LINE Message Data node**
- Type: Code
- JavaScript code to parse incoming JSON to extract `messageId`, `replyToken`, and `messageType` from `body.events[0]`
- Connect from `LINE Webhook` output
3. **Create Get Image from LINE node**
- Type: HTTP Request
- Method: GET
- URL: `https://api-data.line.me/v2/bot/message/{{ $json.messageId }}/content` (dynamic from previous node)
- Response Format: File (binary)
- Headers: `Authorization: Bearer YOUR_TOKEN_HERE` (replace with actual LINE Channel Access Token)
- Connect from `Extract LINE Message Data`
4. **Create OpenAI Vision - Generate Diary node**
- Type: OpenAI (LangChain)
- Operation: Analyze (image)
- Model: `gpt-4o`
- Input Type: Base64 (pass image binary)
- Text prompt: "この写真を見て、50文字以内で日記風の一言を書いてください。"
- Connect from `Get Image from LINE`
5. **Create Prepare Folder Path node**
- Type: Code
- JavaScript builds folder path string `KidsDiary/YYYY/MM` based on current date
- Reads diary text from previous node output (`content`)
- Also collects binary image data from `Get Image from LINE`
- Connect from `OpenAI Vision - Generate Diary`
6. **Create Create Folder Structure node**
- Type: Google Drive (Folder)
- Folder Name: `KidsDiary_{{ $json.year }}_{{ $json.month }}`
- Drive: My Drive
- Parent Folder: root
- Connect from `Prepare Folder Path`
7. **Create Merge node**
- Type: Merge (Combine mode)
- Combine inputs by position (two inputs)
- Connect first input from `Create Folder Structure`
- Connect second input from `Prepare Folder Path` (parallel stream)
8. **Create Convert to File node**
- Type: Convert To File
- Operation: toText
- Source Property: `diaryText` (from merged input)
- Connect from `Merge`
9. **Create Upload Photo to diary　new node**
- Type: Google Drive (File Upload)
- File Name: `{{ $now.format('yyyy-MM-dd_HHmmss') }}_diary.txt` (timestamped)
- Drive: My Drive
- Folder ID: set dynamically from folder creation node output (`$json.id`)
- Connect from `Convert to File`
10. **Create Upload Photo to Drive node**
- Type: Google Drive (File Upload)
- File Name: `{{ $json.diaryText }}` (AI diary sentence as filename)
- Drive: My Drive
- Folder ID: root (note: consider changing to folder ID for logical consistency)
- Connect from `Merge`
11. **Create Reply to LINE node**
- Type: HTTP Request
- Method: POST
- URL: `https://api.line.me/v2/bot/message/reply`
- Headers:
- Authorization: Bearer YOUR_TOKEN_HERE (replace with LINE Channel Access Token)
- Content-Type: application/json
- Body (JSON):
```json
{
"replyToken": "{{ $('Extract LINE Message Data').first().json.replyToken }}",
"messages": [
{
"type": "text",
"text": "写真と日記を保存しました！"
}
]
}
```
- Connect from both `Upload Photo to diary new` and `Upload Photo to Drive` (parallel)
12. **Set credentials:**
- For Google Drive nodes: Configure OAuth2 credentials with appropriate permissions to create folders and upload files
- For HTTP Request nodes calling LINE API: Use a valid LINE Channel Access Token with messaging API permissions
- For OpenAI node: Set up OpenAI API credentials with access to GPT-4o and Vision features
13. **Testing & Validation:**
- Test with real LINE messages containing photos
- Confirm folder creation and file uploads in Google Drive under `KidsDiary/YYYY/MM`
- Confirm reply messages appear in LINE chat
---
### 5. General Notes & Resources
| Note Content | Context or Link |
|----------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------|
| The workflow assumes valid tokens for LINE Messaging API and Google Drive OAuth2 credentials are set up properly. | Required for HTTP Request and Google Drive nodes respectively |
| OpenAI Vision usage requires API access to GPT-4o with Vision capabilities; ensure API subscription supports this. | OpenAI documentation for GPT-4o Vision: https://platform.openai.com/docs/models/gpt-4o |
| Folder naming convention uses underscores and date formatting to keep files organized for easy retrieval later. | Naming pattern: `KidsDiary_YYYY_MM` and diary files timestamped `yyyy-MM-dd_HHmmss_diary.txt` |
| The workflow does not currently handle multiple events in a single webhook call; only processes the first event. | Enhancement could include looping over all events in the webhook payload |
| Consider verifying LINE webhook signature for security to reject unauthorized requests. | LINE docs: https://developers.line.biz/en/docs/messaging-api/verify-signature/ |
| The original photo upload node uses the diary text as filename; this may cause issues if diary text contains invalid filename characters. | Adding sanitization or alternative naming recommended |
| Google Drive "Upload Photo to Drive" node uses root folder ID; consider changing to the created folder for consistency. | Possible bug or oversight in folder ID assignment |
| Sticky notes provide step-by-step documentation visible in the n8n editor to assist users in understanding workflow. | Useful for maintenance and onboarding new users |
---
**Disclaimer:**
The text provided is derived exclusively from an automated workflow created with n8n, a tool for integration and automation. This processing strictly complies with current content policies and contains no illegal, offensive, or protected material. All data handled is legal and public.