Bear Brown
← All N8N workflows
AI & researchAgentExtract From FileFilterHTTP RequestLM Chat Google Gemini

When clicking "Test workflow"

Configured with Agent, Extract From File, Filter, HTTP Request, and additional nodes. Review the connections and setup before adapting this template.

Download workflow JSONView files on GitHub ↗22 nodes · MIT
Workflow preview
TriggerWhen clicking ‘Testworkflow’Manual TriggerAI & modelsGoogle Gemini ChatModelLM Chat Google GeminiAI & modelsAI AgentAgentWeb / APIGet image fromunsplash2HTTP RequestTransformSplit OutSplit OutWeb / APIGet image fromunsplash3HTTP RequestTransformTransform to baseExtract From FileWeb / APICall Gemini API1HTTP RequestRoutingLoop Over ItemsLoop Over ItemsAI & modelsAI Agent2AgentAI & modelsGoogle Gemini ChatModel1LM Chat Google GeminiWeb / APIGet image fromunsplash4HTTP RequestWeb / APIGet PDF fileHTTP RequestWeb / APIGet image from unsplashHTTP RequestWeb / APICall Gemini API withPDFHTTP RequestWeb / APICall Gemini API withImageHTTP RequestTransformTransform to base64(image)Extract From FileTransformTransform to base64(pdf)Extract From FileTransformDefine Multiple ImageURLsEdit FieldsTransformSplit Out to multipleitemsSplit OutTransformDefine URLs And PromptsEdit FieldsRoutingFilter (optional)Filter

Node categories in this workflow

  • Trigger
  • AI & models
  • Web / API
  • Transform
  • Routing

Icons describe general node roles, inferred from node types. Inspect the parameters for the configured action.

Drag to pan; use + and − to zoom. Reconstructed from the exported node positions; nothing runs here. Dashed lines represent non-main connections. Full node names and notes appear below.

How it works

The exported workflow contains the following configured nodes. Their connections are shown above; inspect the JSON for the complete logic, expressions, and external actions.

Workflow node names, types, and operations
NodeCategoryTypeOperation
When clicking ‘Test workflow’TriggerManual TriggerSee node parameters
Google Gemini Chat ModelAI & modelsLM Chat Google GeminiSee node parameters
AI AgentAI & modelsAgentSee node parameters
Get image from unsplash2Web / APIHTTP RequestSee node parameters
Split OutTransformSplit OutSee node parameters
Get image from unsplash3Web / APIHTTP RequestSee node parameters
Transform to baseTransformExtract From FilebinaryToPropery
Call Gemini API1Web / APIHTTP RequestPOST
Loop Over ItemsRoutingLoop Over ItemsSee node parameters
AI Agent2AI & modelsAgentSee node parameters
Google Gemini Chat Model1AI & modelsLM Chat Google GeminiSee node parameters
Get image from unsplash4Web / APIHTTP RequestSee node parameters
Get PDF fileWeb / APIHTTP RequestSee node parameters
Get image from unsplashWeb / APIHTTP RequestSee node parameters
Call Gemini API with PDFWeb / APIHTTP RequestPOST
Call Gemini API with ImageWeb / APIHTTP RequestPOST
Transform to base64 (image)TransformExtract From FilebinaryToPropery
Transform to base64 (pdf)TransformExtract From FilebinaryToPropery
Define Multiple Image URLsTransformEdit FieldsSee node parameters
Split Out to multiple itemsTransformSplit OutSee node parameters
Define URLs And PromptsTransformEdit FieldsSee node parameters
Filter (optional)RoutingFilterSee node parameters

Setup

  1. Download the JSON and import it into an isolated n8n instance. Leave the workflow inactive.
  2. Connect your own credentials through n8n’s credential store. Replace REPLACE_WITH_* values with your resource identifiers.
  3. Inspect API destinations, Code nodes, database queries, file paths, and outbound actions.
  4. Run with synthetic data and harmless destinations. Enable triggers only after checking the actual outputs.

Requirements

An n8n instance with the listed node types and compatible versions. Service-connected workflows may require accounts, credentials, or paid APIs.

Credential types declared in the source

  • googlePalmApi
  • httpQueryAuth

Workflow notes

When clicking "Test workflow"

When clicking "Test workflow"

This trigger demonstrates five different approaches to analyze media with AI:

  1. Top branch: Single image with automatic binary passthrough
  2. Second branch: Multiple images with custom prompts
  3. Third branch: Standard n8n item processing with direct API
  4. Fourth branch: PDF analysis via direct API
  5. Fifth branch: Image analysis via direct API

Each approach has advantages depending on your use case.

METHOD 1: Single image with automatic binary passthrough

METHOD 1: Single image with automatic binary passthrough

This branch demonstrates the easiest way to analyze a single image with AI:

  1. Fetch an image from Unsplash
  2. Send directly to the AI Agent with "Automatically Passthrough Binary Images" enabled
  3. Get AI analysis without any data transformation

BEST FOR: Quick implementation with minimal configuration for single image analysis.

METHOD 2: Multiple images with custom prompts

METHOD 2: Multiple images with custom prompts

This branch shows how to analyze different images with custom instructions:

  1. Prepare data structure with image URLs and their corresponding prompts
  2. Split into individual items and filter if needed
  3. Fetch each image from Unsplash
  4. Process sequentially through the Loop node
  5. Analyze each with its specific prompt using the AI Agent

BEST FOR: When you need different analysis goals for each image (e.g., one for object detection, another for scene description).

METHOD 3: Standard n8n item processing with direct API

METHOD 3: Standard n8n item processing with direct API

This branch demonstrates n8n's standard approach to handling multiple items:

  1. Define multiple image URLs in a single node
  2. Split into individual items for processing
  3. Fetch each image individually
  4. Transform each to base64 format
  5. Make direct API calls to Gemini for each item

BEST FOR: Processing multiple images using n8n's standard item-by-item approach with direct API control.

METHOD 4: PDF analysis via direct API

METHOD 4: PDF analysis via direct API

This branch shows how to analyze PDF documents:

  1. Fetch a PDF file
  2. Transform to base64 format
  3. Send directly to Gemini API for analysis

BEST FOR: Document analysis, text extraction, summarization of PDFs.

METHOD 5: Image analysis via direct API

METHOD 5: Image analysis via direct API

This branch demonstrates direct API control for image analysis:

  1. Fetch an image
  2. Transform to base64 format
  3. Make a customized API call to Gemini

BEST FOR: Advanced users who need precise control over API parameters and response handling.

When clicking "Test workflow"

When clicking "Test workflow"

Purpose and structure

This folder contains the When clicking "Test workflow" workflow. Its executable nodes use agent, extractFromFile, filter, httpRequest, lmChatGoogleGemini, manualTrigger, set, splitInBatches, splitOut. The table and connection map below are derived from the JSON, not an inferred guarantee of behavior.

Setup

Import inactive. Configure your own credentials and replace REPLACE_WITH_* values. Review external destinations and Code nodes before running.

Entry points

When clicking ‘Test workflow’

Verification

Sanitized source candidate; not runtime-tested. This overview is generated from node metadata. Read the folder README for the connection map, credential types, and license.

Attribution

https://github.com/Zie619/n8n-workflows/blob/94007c1445d9258a7da116646b79473e7c7c3282/workflows/Splitout/0772_Splitout_Filter_Process_Webhook.json MIT; preserve source notices.

More in AI & research

"Hey Siri, Ask Agent" workflow#️⃣Nostr #damus AI Powered Reporting + Gmail + Telegram(G) - Email Classification