Capture Lightroom to Instagram Queue with AI Alt Text - n8n Workflow

• Professional photographers managing their workflow from Adobe Lightroom Cloud.


• Social media managers needing an automated queue of visual content.


• Developers looking for advanced n8n templates integrating custom API calls and LLMs.


• Users aiming to streamline content creation by generating accessibility-focused alt text automatically.

This complex and powerful n8n workflow solves the tedious task of manually tracking new photos ready for social media posting and writing descriptive alt text for accessibility and SEO.

The system works by using an n8n trigger (Schedule Trigger) to periodically poll a specific Adobe Lightroom album. Because the Adobe API response often includes a non-standard XSSI prefix, the workflow utilizes a custom JavaScript n8n node to clean the data before processing. Once cleaned, the system checks the asset against an n8n Data Table to prevent duplicate processing.

For new assets, the n8n workflow uses an Anthropic (Claude) AI agent to analyze the image content (retrieved via a secure image proxy webhook URL) and generate high-quality, SEO-optimized alt descriptions. Finally, the image metadata and the newly generated alt text are stored in the n8n Data Table, creating a perfect queue for a secondary posting n8n workflow.

This automation initiates with an n8n trigger and follows a structured flow:

1. Schedule Trigger: The n8n workflow starts on a set schedule (e.g., hourly polling).


2. Configuration Setting: A Set n8n node (Params) injects critical credentials and IDs (Lightroom catalog ID, album ID, and n8n instance URL) into the data stream.


3. Fetch Lightroom Assets: An HTTP Request n8n node connects to the Adobe Lightroom API, retrieving all assets from the target album using OAuth credentials.


4. Clean Data: A Code in JavaScript n8n node strips the while(1){} XSSI prefix from the raw API response, ensuring the data can be reliably parsed.


5. Split and Sort: The workflow splits the collection of resources into individual items and sorts them, typically by capture date, using the Sort n8n node.


6. Deduplication Check: A Data Table n8n node (If row does not exist) checks if the asset's ID already exists in the 'Photos' data table. Only unique, new items proceed.


7. AI Analysis (Batch Loop): New assets enter a loop. The Anthropic n8n node retrieves the image via the calculated webhook proxy URL and uses a specialized prompt to generate an accessible and descriptive alt text.


8. Store in Queue: The final step is an Insert row Data Table n8n node, which saves the asset metadata and the generated alt text into the 'Photos' table. This table acts as the posting queue for subsequent n8n templates.

1. Import: Copy the provided n8n workflow JSON and paste it into your n8n instance.


2. Credentials Setup: Configure the required credentials:

1. Data Table: Create a new n8n Data Table named 'Photos' with at least the columns specified in the sticky note: lrassetid (used for deduplication), lr_asset, and alt.


2. Configure Params Node: Update the Params n8n node with your specific Lightroom Catalog ID, Album ID, API Key (if not using shared credentials), and, critically, the public base URL of your n8n instance (used for the image proxy webhook in the AI analysis step).


3. Webhook Configuration: Ensure your n8n instance is publicly accessible if you plan to use the image proxy webhook URL referenced in the Analyze image n8n node.


4. Activate: Save and activate the n8n workflow.

• Schedule Trigger: The starting n8n trigger, configured to run periodically (e.g., every hour) to poll the Lightroom API.


• Params (Set): A setup n8n node used to centralize and inject configuration parameters like API keys, catalog IDs, and the n8n instance URL. Essential for dynamic URL construction in later steps.


• HTTP Request: Fetches the asset list from the Adobe Lightroom API using customized headers (X-API-Key) and OAuth authentication.


• Code in JavaScript: A crucial custom code n8n node dedicated to data cleanup. It strips the leading while(1){} XSSI defense prefix from the Adobe API text response so that the data can be parsed into valid JSON.


• Split Out: Breaks the single API response containing multiple assets into individual items for per-photo processing.


• If row does not exist (Data Table): Implements deduplication logic. This n8n node checks the 'Photos' Data Table to see if the current lrassetid has already been processed and queued, preventing unnecessary AI analysis and duplicate entries.


• Analyze image (Anthropic): The core AI n8n node. It analyzes the image URL (fetched via a dynamic webhook endpoint) and uses a sophisticated prompt to generate a descriptive alt text string, using the Claude model.


• Insert row (Data Table): The final action n8n node, which stores the original asset data and the newly generated AI alt text into the 'Photos' queue table.