agente
Configured with Agent, Convert To File, Evolution Api, Google Tasks Tool, and additional nodes. Review the connections and setup before adapting this template.
Node categories in this workflow
- AI & models
- Integration
- Trigger
- Data store
- Messaging
- Transform
- Routing
- Subworkflow
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.
| Node | Category | Type | Operation |
|---|---|---|---|
| OpenAI Chat Model1 | AI & models | OpenAI Chat Model | See node parameters |
| Assistente clinica interno | AI & models | Agent | See node parameters |
| Google Tasks | Integration | Google Tasks Tool | See node parameters |
| MCP Google Calendar | Integration | Mcp Client Tool | See node parameters |
| Receber Mensagem Telegram | Trigger | Telegram Trigger | See node parameters |
| Postgres Chat Memory | Data store | Memory Postgres Chat | See node parameters |
| OpenAI Chat Model | AI & models | OpenAI Chat Model | See node parameters |
| Assistente Clínica | AI & models | Agent | See node parameters |
| Enviar alerta de cancelamento | Messaging | Telegram Tool | See node parameters |
| Gatilho diário | Trigger | Schedule Trigger | See node parameters |
| Assistente de confirmação | AI & models | Agent | See node parameters |
| AI Agent | AI & models | Agent | See node parameters |
| Postgres Chat Memory1 | Data store | Memory Postgres Chat | See node parameters |
| MCP Google Calendar2 | Integration | Mcp Client Tool | See node parameters |
| MCP GMAIL | Integration | Mcp Client Tool | See node parameters |
| MCP CALENDAR | Integration | Mcp Client Tool | See node parameters |
| Telegram | Messaging | Telegram | See node parameters |
| OpenRouter Chat Model1 | AI & models | LM Chat Open Router | See node parameters |
| Webhook1 | Trigger | Webhook | See node parameters |
| Edit Fields1 | Transform | Edit Fields | See node parameters |
| Switch | Routing | Switch | See node parameters |
| OpenAI | AI & models | Open Ai | audio / transcribe |
| Evolution API | Integration | Evolution Api | chat-api / get-media-base64 |
| Convert to File | Transform | Convert To File | toBinary |
| OpenAI1 | AI & models | Open Ai | image / analyze |
| AI Agent2 | AI & models | Agent | See node parameters |
| OpenRouter Chat Model2 | AI & models | LM Chat Open Router | See node parameters |
| Evolution API2 | Integration | Evolution Api | messages-api |
| CallToHuman | Subworkflow | Tool Workflow | See node parameters |
| OpenAI Chat Model2 | AI & models | OpenAI Chat Model | See node parameters |
| REMINDER | Integration | Evolution Api | messages-api |
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
evolutionApigoogleTasksOAuth2ApiopenAiApiopenRouterApipostgrestelegramApi
Workflow notes
"Appointment Confirmation Assistant"
"Appointment Confirmation Assistant"
Description:
Purpose: This section contains the configuration for the Appointment Confirmation Assistant, an agent specialized in confirming scheduled appointments with patients.
Instructions for Use:
It is triggered automatically every weekday (Monday to Friday) at 08:00 AM via the Daily Trigger (Gatilho diário).
The agent retrieves all appointments scheduled for the next day using MCP Google Calendar.
It extracts each patient's phone number from the event description field.
A confirmation message is sent to each patient using the relembraAGENDAMENTO tool, asking for confirmation or rescheduling.
Important: This agent does not handle responses from patients; another agent or workflow is responsible for follow-ups.
Make sure event descriptions in Google Calendar are correctly filled to avoid errors.
"Agent Core Components (Tools, MCP, Memory, LLM Model)"
"Agent Core Components (Tools, MCP, Memory, LLM Model)"
Description:
Purpose:
This sticky note represents the essential structure of any intelligent agent: it includes access to external tools, persistent memory, the MCP system for calendar management, and a Language Model (LLM) to process natural language tasks.
Instructions for Use:
Language Model nodes (OpenAI, OpenRouter) are responsible for natural language understanding and generation.
Memory nodes (Postgres Chat Memory) maintain conversation context over multiple interactions.
MCP Tools interact with Google Calendar and other services to perform real-world actions.
Always ensure memory synchronization between the agent's context and actions performed.
If new tools are added, they must be connected both to the agent and properly described in the system message.
"Processing and Sending WhatsApp Responses"
"Processing and Sending WhatsApp Responses"
Description:
Purpose: This section is responsible for processing, formatting, and sending outbound WhatsApp messages to patients through the Evolution API.
Instructions for Use:
Messages received from the assistant agent are first reformatted by the AI Agent node to comply with WhatsApp markdown syntax (e.g., replacing bold with bold).
Once formatted, the messages are forwarded to WhatsApp using the Evolution API2 node.
Ensure proper formatting before sending to maintain a professional communication tone and avoid delivery errors.
Any future text-processing improvements should be implemented here.
Incoming WhatsApp Webhook and Message Type Handling"
Incoming WhatsApp Webhook and Message Type Handling"
Description:
Purpose: Manages the initial reception and classification of incoming WhatsApp messages from patients via the webhook system.
Instructions for Use:
The Webhook1 node captures incoming messages.
Edit Fields1 extracts structured fields such as text, image URL, audio URL, and document URL.
Switch node analyzes which type of content was received and directs the flow accordingly:
Text → Forwarded to the assistant for handling.
Image → Sent for OCR analysis.
Audio → Sent for transcription.
Document → (Currently unused, but ready for future workflows).
Keep webhook credentials updated to ensure system reliability.
"Download Audio and Convert to MP4"
"Download Audio and Convert to MP4"
Description:
Purpose: Handles retrieval, conversion, and transcription of audio files sent by patients via WhatsApp.
Instructions for Use:
Evolution API downloads the audio in base64 format.
Convert to File transforms base64 into a binary file compatible with transcription engines.
OpenAI Whisper API (via OpenAI node) transcribes the audio into text, preparing it for natural language processing.
Ensure audio formats are correctly handled (e.g., MP4/MP3) to avoid conversion or transcription failures.
Monitor for potential heavy file size issues (>25MB) which may impact performance.
"Extract Text from Images"
"Extract Text from Images"
Description:
Purpose: Processes images received via WhatsApp to extract text (OCR) and describe their visual content for further contextual analysis.
Instructions for Use:
The OpenAI1 node uses a Vision model to transcribe and describe any text or scene from incoming images.
The resulting data is interpreted by the AI Agent2, which prepares insights and recommends appropriate responses.
Image-to-text extraction is especially useful for handling prescriptions, notes, or medical documents sent by patients.
Maintain quality standards: images must be clear and high-resolution for best transcription results.
"Internal Clinic Assistant"
"Internal Clinic Assistant"
Description:
Purpose: Represents the Internal Assistant for the clinic, used exclusively by the internal team via Telegram to manage patient rescheduling and manage a purchasing reminder list.
Instructions for Use:
Triggered by staff messages sent via Telegram.
Rescheduling tasks:
Access the MCP Google Calendar to locate and manage appointments.
Extract patient contact from the event description.
Send rescheduling messages through WhatsApp using the dedicated tool.
Shopping list tasks:
Insert shopping list reminders into Google Tasks based on the staff's instructions.
Always maintain professional and empathetic tone when interacting with patients, even if the communication is indirect.
Avoid unauthorized direct patient contact without staff permission.
agente
agente
Purpose and structure
This folder contains the agente workflow. Its executable nodes use agent, convertToFile, evolutionApi, googleTasksTool, lmChatOpenAi, lmChatOpenRouter, mcpClientTool, memoryPostgresChat, openAi, scheduleTrigger, set, switch, telegram, telegramTool, telegramTrigger, toolWorkflow, webhook. 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
Receber Mensagem Telegram, Gatilho diário, Webhook1
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/Googletaskstool/1103_Googletaskstool_Telegram_Automation_Webhook.json MIT; preserve source notices.