AI & researchAggregateCodeDocument Default Data LoaderEmbeddings Google GeminiExecute Workflow

Stage 1 - Research for API Documentation

Configured with Aggregate, Document Default Data Loader, Embeddings Google Gemini, Execute Workflow, and additional nodes. Review the connections and setup before adapting this template.

Download workflow JSON View files on GitHub ↗82 nodes · MIT

Workflow preview

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
Node	Category	Type	Operation
When clicking ‘Test workflow’	Trigger	Manual Trigger	See node parameters
Web Search For API Schema	Web / API	HTTP Request	POST
Scrape Webpage Contents	Web / API	HTTP Request	See node parameters
Results to List	Transform	Split Out	See node parameters
Recursive Character Text Splitter1	Transform	Text Splitter Recursive Character Text Splitter	See node parameters
Content Chunking @ 50k Chars	Transform	Edit Fields	See node parameters
Split Out Chunks	Transform	Split Out	See node parameters
Default Data Loader	Files / media	Document Default Data Loader	See node parameters
Set Embedding Variables	Transform	Edit Fields	See node parameters
Execute Workflow Trigger	Trigger	Execute Workflow Trigger	See node parameters
Execution Data	Integration	Execution Data	See node parameters
EventRouter	Routing	Switch	See node parameters
Google Gemini Chat Model	AI & models	LM Chat Google Gemini	See node parameters
Successful Runs	Routing	Filter	See node parameters
For Each Document...	Routing	Loop Over Items	See node parameters
Embeddings Google Gemini	AI & models	Embeddings Google Gemini	See node parameters
Has API Documentation?	AI & models	Text Classifier	See node parameters
Store Document Embeddings	Data store	Vector Store Qdrant	See node parameters
Embeddings Google Gemini1	AI & models	Embeddings Google Gemini	See node parameters
Google Gemini Chat Model1	AI & models	LM Chat Google Gemini	See node parameters
Extract API Operations	AI & models	Information Extractor	See node parameters
Search in Relevant Docs	Data store	Vector Store Qdrant	See node parameters
Wait	Wait	Wait	See node parameters
Remove Dupes	Transform	Remove Duplicates	See node parameters
Filter Results	Routing	Filter	See node parameters
Research	Subworkflow	Execute Workflow	See node parameters
Has Results?	Routing	If	See node parameters
Response Empty	Transform	Edit Fields	See node parameters
Response OK	Transform	Edit Fields	See node parameters
Combine Docs	Transform	Aggregate	See node parameters
Template to List	Transform	Split Out	See node parameters
Query Templates	Transform	Edit Fields	See node parameters
Google Gemini Chat Model2	AI & models	LM Chat Google Gemini	See node parameters
For Each Template...	Routing	Loop Over Items	See node parameters
Query & Docs	Transform	Edit Fields	See node parameters
Identify Service Products	AI & models	Information Extractor	See node parameters
Extract API Templates	Transform	Edit Fields	See node parameters
Embeddings Google Gemini2	AI & models	Embeddings Google Gemini	See node parameters
Search in Relevant Docs1	Data store	Vector Store Qdrant	See node parameters
Combine Docs1	Transform	Aggregate	See node parameters
Query & Docs1	Transform	Edit Fields	See node parameters
For Each Template...1	Routing	Loop Over Items	See node parameters
Merge Lists	Code	Code	See node parameters
Remove Duplicates	Transform	Remove Duplicates	See node parameters
Append Row	Data store	Google Sheets	append
Response OK1	Transform	Edit Fields	See node parameters
Has Operations?	Routing	If	See node parameters
Response Empty1	Transform	Edit Fields	See node parameters
Research Pending	Data store	Google Sheets	update
Research Result	Data store	Google Sheets	update
Research Error	Data store	Google Sheets	update
Extract Pending	Data store	Google Sheets	update
Research Event	Transform	Edit Fields	See node parameters
Extract Event	Transform	Edit Fields	See node parameters
Extract	Subworkflow	Execute Workflow	See node parameters
Extract Result	Data store	Google Sheets	update
Extract Error	Data store	Google Sheets	update
Get API Operations	Data store	Google Sheets	See node parameters
Contruct JSON Schema	Code	Code	See node parameters
Upload to Drive	Files / media	Google Drive	createFromText
Set Upload Fields	Transform	Edit Fields	See node parameters
Response OK2	Transform	Edit Fields	See node parameters
Generate Event	Transform	Edit Fields	See node parameters
Generate Pending	Data store	Google Sheets	update
Generate	Subworkflow	Execute Workflow	See node parameters
Generate Error	Data store	Google Sheets	update
Generate Result	Data store	Google Sheets	update
Get All Extract	Data store	Google Sheets	See node parameters
Get All Research	Data store	Google Sheets	See node parameters
For Each Research...	Routing	Loop Over Items	See node parameters
For Each Extract...	Routing	Loop Over Items	See node parameters
Wait1	Wait	Wait	See node parameters
All Research Done?	Routing	If	See node parameters
All Extract Done?	Routing	If	See node parameters
Get All Generate	Data store	Google Sheets	See node parameters
All Generate Done?	Routing	If	See node parameters
For Each Generate...	Routing	Loop Over Items	See node parameters
Wait2	Wait	Wait	See node parameters
Has Results?1	Routing	If	See node parameters
Response Scrape Error	Transform	Edit Fields	See node parameters
Has Results?3	Routing	If	See node parameters
Response No API Docs	Transform	Edit Fields	See node parameters

Setup

Download the JSON and import it into an isolated n8n instance. Leave the workflow inactive.
Connect your own credentials through n8n’s credential store. Replace REPLACE_WITH_* values with your resource identifiers.
Inspect API destinations, Code nodes, database queries, file paths, and outbound actions.
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

googleSheetsOAuth2Api

Workflow notes

Stage 1 - Research for API Documentation

Fetch a list of services pending research from Database (Google Sheet)
Uses a search engine (Google) to find API Documentation for each service
Uses Webscraper (Apify) to read the contents of search results to filter irrelevant pages
Stores webpage contents and metadata into Vector Store (Qdrant)

Stage 2 - Extract API Operations From Documentation

Fetch a list of services pending extraction from Database (Google Sheet)
Query Vector store (Qdrant) to figure out service's products, solutions and offerings
Query Vector store (Qdrant) again for API documentation relevant to these products, solutions and offerings
Extract any API operations found in the API documentation results using LLM (Gemini)
Store extracted API operations into Database (Google Sheet)

Stage 3 - Generate Custom Schema From API Operations

Fetch a list of services pending generation from Database (Google Sheet)
Fetch all API operations for each service from Database (Google Sheet)
Use Code node to combine and group all API operations for a service and convert to a custom schema
Upload the resulting custom schema to file storage (Google Drive)

Stage 1 - Research for API Documentation

Purpose and structure

This folder contains the Stage 1 - Research for API Documentation workflow. Its executable nodes use aggregate, code, documentDefaultDataLoader, embeddingsGoogleGemini, executeWorkflow, executeWorkflowTrigger, executionData, filter, googleDrive, googleSheets, httpRequest, if, informationExtractor, lmChatGoogleGemini, manualTrigger, removeDuplicates, set, splitInBatches, splitOut, switch, textClassifier, textSplitterRecursiveCharacterTextSplitter, vectorStoreQdrant, wait. 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’, Execute Workflow Trigger

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/Wait/0668_Wait_Splitout_Create_Webhook.json MIT; preserve source notices.

Canvas section labels

These name areas of the original canvas; they do not include explanatory instructions.

Stage 1 - Subworkflow
Stage 2 - Subworkflow
Stage 3 - Subworkflow

More in AI & research

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

Node categories in this workflow

How it works

Setup

Requirements

Credential types declared in the source

Workflow notes