January 21, 2026

Telegram + Airtable: an assistant that remembers

Lisa Granqvist Partner Workflow Automation Expert

Get a free AI assessment → ⬇️ Use template

Your “assistant” probably doesn’t remember anything. You tell it a preference, a decision, or a task… and two days later you’re repeating yourself in a different app, hunting through chat history, or guessing what you meant.

This is the kind of mess that slows down marketers trying to move fast, and it drives agency owners nuts when context gets scattered across clients. Even a busy founder feels it. A Telegram Airtable assistant turns those one-off chats into something reliable, with memory you can actually use.

This workflow gives you a Telegram-based AI agent that remembers key facts in Airtable, transcribes voice notes, and pulls answers from Gmail, Google Calendar, Notion, and web search. You’ll see what it automates, what results to expect, and what you need to run it smoothly.

How This Automation Works

The full n8n workflow, from trigger to final output:

n8n Workflow Template: Telegram + Airtable: an assistant that remembers

Click to explore

flowchart LR

    subgraph sg0["Telegram Incoming Hook Flow"]
        direction LR
        n0@{ icon: "mdi:memory", form: "rounded", label: "Brief Memory Buffer", pos: "b", h: 48 }
        n1@{ icon: "mdi:database", form: "rounded", label: "Store Memory Record", pos: "b", h: 48 }
        n2["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/airtable.svg' width='40' height='40' /></div><br/>Retrieve Memory Entries"]
        n3@{ icon: "mdi:cog", form: "rounded", label: "Compile Memory List", pos: "b", h: 48 }
        n4["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/merge.svg' width='40' height='40' /></div><br/>Combine Inputs"]
        n5["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/telegram.svg' width='40' height='40' /></div><br/>Telegram Incoming Hook"]
        n6["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/telegram.svg' width='40' height='40' /></div><br/>Send Telegram Reply"]
        n7@{ icon: "mdi:wrench", form: "rounded", label: "Web Search via SerpAPI", pos: "b", h: 48 }
        n8@{ icon: "mdi:location-exit", form: "rounded", label: "Fetch Calendar Events", pos: "b", h: 48 }
        n9@{ icon: "mdi:location-exit", form: "rounded", label: "Create Calendar Event", pos: "b", h: 48 }
        n10@{ icon: "mdi:location-exit", form: "rounded", label: "Modify Calendar Event", pos: "b", h: 48 }
        n11@{ icon: "mdi:location-exit", form: "rounded", label: "Remove Calendar Event", pos: "b", h: 48 }
        n12@{ icon: "mdi:message-outline", form: "rounded", label: "Retrieve Gmail Threads", pos: "b", h: 48 }
        n13@{ icon: "mdi:message-outline", form: "rounded", label: "Fetch Gmail Message", pos: "b", h: 48 }
        n14@{ icon: "mdi:brain", form: "rounded", label: "Grok Chat Model", pos: "b", h: 48 }
        n15@{ icon: "mdi:swap-horizontal", form: "rounded", label: "Text or Voice Check", pos: "b", h: 48 }
        n16@{ icon: "mdi:robot", form: "rounded", label: "Transcribe Audio Clip", pos: "b", h: 48 }
        n17["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/telegram.svg' width='40' height='40' /></div><br/>Download Voice File"]
        n18@{ icon: "mdi:database", form: "rounded", label: "Fetch Notion Database", pos: "b", h: 48 }
        n19@{ icon: "mdi:robot", form: "rounded", label: "Conversation Orchestrator", pos: "b", h: 48 }
        n19 --> n6
        n15 --> n4
        n15 --> n2
        n15 --> n17
        n4 --> n19
        n7 -.-> n19
        n3 --> n4
        n17 --> n16
        n1 -.-> n19
        n2 --> n3
        n0 -.-> n19
        n5 --> n15
        n14 -.-> n19
        n13 -.-> n19
        n16 --> n4
        n16 --> n2
        n18 -.-> n19
        n12 -.-> n19
        n9 -.-> n19
        n11 -.-> n19
        n8 -.-> n19
        n10 -.-> n19
    end

    %% Styling
    classDef trigger fill:#e8f5e9,stroke:#388e3c,stroke-width:2px
    classDef ai fill:#e3f2fd,stroke:#1976d2,stroke-width:2px
    classDef aiModel fill:#e8eaf6,stroke:#3f51b5,stroke-width:2px
    classDef decision fill:#fff8e1,stroke:#f9a825,stroke-width:2px
    classDef database fill:#fce4ec,stroke:#c2185b,stroke-width:2px
    classDef api fill:#fff3e0,stroke:#e65100,stroke-width:2px
    classDef code fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px
    classDef disabled stroke-dasharray: 5 5,opacity: 0.5
    class n5 trigger
    class n16,n19 ai
    class n14 aiModel
    class n7 ai
    class n0 ai
    class n15 decision
    class n1,n2,n18 database
    classDef customIcon fill:none,stroke:none
    class n2,n4,n5,n6,n17 customIcon

The Problem: Your “Assistant” Has No Memory (So You Do the Work Twice)

Most teams don’t struggle because they lack tools. They struggle because their tools don’t share context. You make a plan in chat, confirm a time in email, save a link in Notion, then try to stitch it all together later when you actually need it. That’s when mistakes sneak in. Calendar events get created without the right details, follow-ups get missed, and you waste about 30 minutes just reloading your own brain every morning.

It adds up fast. Here’s where it breaks down in real life.

You repeat preferences and “important facts” constantly, because chat history is not a database.
Finding the right email thread means leaving Telegram, opening Gmail, searching, clicking, and losing your train of thought.
Calendar admin becomes a mini project, especially when you’re capturing tasks from voice notes and trying to convert them into real events.
Notes and reading lists sit in Notion, but you rarely recall them when you need them most.

The Solution: A Telegram Agent With Airtable Memory + On-Demand Tools

This n8n workflow turns Telegram into a front door for work you normally do across five tabs. You message your bot (text or voice), and the automation routes the request, pulls your recent Airtable “memories” for context, and hands everything to an AI agent. If you send a voice note, it automatically downloads the audio, transcribes it with OpenAI, then treats it like a normal message. From there the agent can decide to search the web, read Gmail, manage Google Calendar events, or fetch a Notion database, depending on what you asked. When you share a personal preference or recurring detail, it quietly saves a distilled version into Airtable so your future requests feel less repetitive.

The workflow starts when a Telegram message hits your bot. It then loads your latest Airtable memories and (if needed) transcribes audio. Finally, the agent combines that context with live tool outputs and replies right back in Telegram, often with the task already done.

What You Get: Automation vs. Results

What This Workflow Automates

Results You’ll Get

Routes Telegram messages as text or voice and transcribes voice notes automatically.
Pulls your most recent Airtable memory entries for the same Telegram user.
Lets the agent call Gmail, Google Calendar, Notion, and web search only when needed.
Saves new personal facts back into Airtable without interrupting the conversation.

Most people get about 30 minutes back per day from less tab switching.
Fewer “wait, what did we decide?” moments because preferences and decisions persist.
Voice-first capture becomes usable, so quick notes don’t die in chat.
Calendar updates happen in the same thread where the request was made.
Your Notion info is easier to retrieve, which makes it more likely you’ll use it.

Example: What This Looks Like

Say you handle scheduling and follow-ups from chat. Manually, you might spend 10 minutes jumping into Gmail to find context, another 10 minutes creating or updating a Calendar event, then 5 minutes digging through Notion for a related note. That’s about 25 minutes per request, and two requests a day is roughly an hour. With this workflow, you send one Telegram message (or a 30-second voice note), wait a minute for tool calls and transcription, and you’re done in about 5 minutes of active effort.

What You’ll Need

n8n instance (try n8n Cloud free)
Self-hosting option if you prefer (Hostinger works well)
Telegram Bot API to receive and reply to messages
Airtable to store long-term “memories” per user
SerpAPI key (get it from your SerpAPI dashboard)

Skill level: Intermediate. You’ll connect a few accounts, create an Airtable base/table, and paste in a Telegram chat ID.

Don’t want to set this up yourself? Talk to an automation expert (free 15-minute consultation).

How It Works

A Telegram message triggers everything. Your bot receives a text message or voice note via the Telegram Trigger, and the workflow immediately routes it down the right path.

Voice gets turned into usable text. If it’s audio, n8n downloads the file and sends it to OpenAI speech-to-text. You end up with a clean transcript that the agent can reason over, search with, and store as a future reference.

Context is loaded from Airtable and combined. The workflow searches your Airtable “Agent Memories” table for the same user, bundles recent entries into a memory block, and merges that with the current message. There’s also a short-term memory window (about 30 turns) so the agent doesn’t feel forgetful mid-conversation.

The agent uses tools only when they matter. Ask for your upcoming meetings, and it calls Google Calendar. Ask for the latest “Product Updates” email, and it uses Gmail. Need background info, and it hits SerpAPI. Then it replies in Telegram with grounded results and (when appropriate) it silently saves a new memory to Airtable.

You can easily modify which facts get stored to memory based on your needs. See the full implementation guide below for customization options.

Step-by-Step Implementation Guide

Step 1: Configure the Telegram Trigger

Set up the workflow to receive messages from Telegram and pass them into the logic that distinguishes text from voice inputs.

Add and open Telegram Incoming Hook.
Set Updates to message.
Set Additional Fields → Chat IDs to [YOUR_ID] and enable Download to true.
Credential Required: Connect your telegramApi credentials.

Tip: Restricting Chat IDs helps prevent unauthorized access to your assistant.

Step 2: Route Text vs Voice Messages

Route messages so text goes directly into the AI pipeline, and voice messages are downloaded and transcribed first.

Open Text or Voice Check and keep the condition String → Exists with Left Value set to {{ $json.message.text }}.
Connect Text or Voice Check to Download Voice File on the false branch (voice messages).
In Download Voice File, set Resource to file and File ID to {{ $json.message.voice.file_id }}.
Credential Required: Connect your telegramApi credentials in Download Voice File.
In Transcribe Audio Clip, set Resource to audio and Operation to transcribe.
Credential Required: Connect your openAiApi credentials in Transcribe Audio Clip.

⚠️ Common Pitfall: If voice messages aren’t downloading, double-check that Telegram Incoming Hook has Download enabled and the correct Chat IDs.

Step 3: Load and Compile User Memory

Pull past memories from Airtable and aggregate them into a single list for the AI to use.

Open Retrieve Memory Entries and set Operation to search.
Set Filter By Formula to ({User} = '[YOUR_NAME]').
Credential Required: Connect your airtableTokenApi credentials in Retrieve Memory Entries.
Open Compile Memory List and set Fields to Include to Memory, Created and Destination Field Name to Memories.
Open Combine Inputs and set Mode to combine and Combine By to combineAll.

Tip: Text or Voice Check outputs to both Combine Inputs and Retrieve Memory Entries in parallel, keeping responses fast while memory loads.

Step 4: Configure the AI Orchestrator and Memory Tools

Set up the conversation agent, its memory window, and connected tools that the model can call on demand.

Open Conversation Orchestrator and keep Text set to {{ $json.text }}{{ $json.message.text }}.
Confirm the system message in Conversation Orchestrator includes {{ $json.Memories.toJsonString() }} for memory injection.
Open Brief Memory Buffer and set Session Key to {{ $('Text or Voice Check').item.json.message.chat.id }} with Context Window Length set to 30.
Attach Grok Chat Model as the language model for Conversation Orchestrator and set Model to grok-4-0709.
Credential Required: Connect your xAiApi credentials in Grok Chat Model.

⚠️ Common Pitfall: The AI tools and memory components (like Brief Memory Buffer, Store Memory Record, and Web Search via SerpAPI) are attached to Conversation Orchestrator. Ensure their credentials are configured on each tool node, not inside the agent.

Step 5: Connect External Tools for Search, Calendar, Email, and Notion

Enable the assistant to call external tools when users ask for web searches, calendar actions, Gmail queries, or Notion data.

In Web Search via SerpAPI, keep default settings and add credentials. Credential Required: Connect your serpApi credentials.
Configure the Google Calendar tools: Fetch Calendar Events with Operation getAll and Time Max {{ $now.plus({ week: 4 }) }}; Create Calendar Event, Modify Calendar Event, and Remove Calendar Event for event changes. Credential Required: Connect your googleCalendarOAuth2Api credentials to each of these nodes.
Configure Gmail tools: Retrieve Gmail Threads with Operation getAll, and Fetch Gmail Message with Operation get. Keep AI-driven fields like Simplify and Message ID as expressions. Credential Required: Connect your gmailOAuth2 credentials on both nodes.
In Fetch Notion Database, set Resource to database and select your Database ID. Credential Required: Connect your notionApi credentials.
In Store Memory Record, set Airtable Base and Table IDs, and keep Memory mapped to {{ $fromAI('Memory') }}. Credential Required: Connect your airtableTokenApi credentials.

Tip: All tools listed here are connected as AI tools for Conversation Orchestrator, so the agent can call them dynamically based on user requests.

Step 6: Configure the Telegram Response

Return the assistant’s output back to the user in Telegram.

Open Send Telegram Reply and set Text to {{ $json.output }}.
Set Chat ID to {{ $('Telegram Incoming Hook').item.json.message.chat.id }}.
Credential Required: Connect your telegramApi credentials.

Tip: Conversation Orchestrator flows directly into Send Telegram Reply to keep the assistant’s response instantaneous.

Step 7: Test and Activate Your Workflow

Verify the workflow end-to-end and then enable it for live use.

Click Execute Workflow and send a text message to your Telegram bot to trigger Telegram Incoming Hook.
Test a voice note to ensure Download Voice File and Transcribe Audio Clip run before Combine Inputs and Retrieve Memory Entries in parallel.
Confirm you receive a reply in Telegram from Send Telegram Reply with the AI output.
When everything works, toggle the workflow to Active to run continuously.

🔒

Unlock Full Step-by-Step Guide

Get the complete implementation guide + downloadable template

Common Gotchas

Airtable credentials can expire or your base/table names won’t match. If things break, check the Airtable connection in n8n and verify the “Agent Memories” table fields (Memory, User, Created) first.
If you’re using Wait nodes or external rendering, processing times vary. Bump up the wait duration if downstream nodes fail on empty responses.
Default prompts in AI nodes are generic. Add your brand voice early or you’ll be editing outputs forever.

Frequently Asked Questions

How long does it take to set up this Telegram Airtable assistant automation?

About an hour if your accounts and permissions are ready.

Do I need coding skills to automate Telegram Airtable assistant tasks?

No. You’ll mostly be connecting accounts and copying a few IDs into the right fields.

Is n8n free to use for this Telegram Airtable assistant workflow?

Yes. n8n has a free self-hosted option and a free trial on n8n Cloud. Cloud plans start at $20/month for higher volume. You’ll also need to factor in API costs for things like OpenAI transcription and SerpAPI web search.

Where can I host n8n to run this automation?

Two options: n8n Cloud (managed, easiest setup) or self-hosting on a VPS. For self-hosting, Hostinger VPS is affordable and handles n8n well. Self-hosting gives you unlimited executions but requires basic server management.

Can I customize this Telegram Airtable assistant workflow for my brand voice and what it remembers?

Yes, and you should. Update the Agent system message to match your tone, then adjust what gets stored by refining the “Save Memory” tool behavior so it only saves useful, non-sensitive facts. Common tweaks include saving “preferences” but not “one-off plans,” adding categories or tags in Airtable, and shortening the memory window so old context doesn’t pollute new requests.

Why is my Telegram connection failing in this workflow?

Usually it’s the bot token or chat ID. Regenerate the token in @BotFather if needed, and confirm the Telegram Trigger is set to the correct allowed chatIds. Also make sure “download” is enabled if you’re sending voice notes, because the workflow needs permission to fetch the audio file before transcription.

How many messages can this Telegram Airtable assistant automation handle?

It depends on your n8n plan and your API limits.

Is this Telegram Airtable assistant automation better than using Zapier or Make?

Often, yes, if you want a real “agent” with memory and tool-calling. n8n is better suited to branching logic, merging memory context, and mixing multiple tool actions (Calendar, Gmail, Notion, web search) in one conversation without paying per tiny step. It’s also self-hostable, which can matter when volumes grow. Zapier or Make can feel simpler for lightweight, two-app automations, but this workflow is more than a single trigger and action. If you’re torn, Talk to an automation expert and map it to your actual usage.

Once this is running, Telegram stops being a place where requests go to die. Your assistant remembers, your tools respond, and you get to keep moving.

Need Help Setting This Up?

Our automation experts can build and customize this workflow for your specific needs. Free 15-minute consultation—no commitment required.

Lisa Granqvist

Workflow Automation Expert

Expert in workflow automation and no-code tools.

{
"id": null,
"meta": {
"instanceId": "203d410f6b9aca37bc3731484379d8c437e5937524a25f501e625e72b23d9c21",
"templateId": null,
"templateCredsSetupCompleted": null
},
"name": "Telegram-Based AI Assistant Workflow",
"tags": [],
"nodes": [
{
"id": "flowpast-topbar-7452",
"name": "Flowpast Branding",
"type": "n8n-nodes-base.stickyNote",
"position": [
0,
20
],
"parameters": {
"color": 7,
"width": 1505,
"height": 80,
"content": "## Flowpast.com | Automation Workflow Library\n**\ud83d\udcd6 Full tutorial & setup guide:** flowpast.com"
},
"typeVersion": 1
},
{
"id": "15741804-50b9-4252-89e3-0f135c6a127f",
"name": "Brief Memory Buffer",
"type": "@n8n/n8n-nodes-langchain.memoryBufferWindow",
"position": [
270,
125
],
"parameters": {
"sessionKey": "={{ $('Text or Voice Check').item.json.message.chat.id }}",
"sessionIdType": "customKey",
"contextWindowLength": 30
},
"typeVersion": 1.3
},
{
"id": "81ea7f9d-e749-46c7-be40-616429a439e7",
"name": "Store Memory Record",
"type": "n8n-nodes-base.airtableTool",
"position": [
95,
155
],
"parameters": {
"base": {
"__rl": true,
"mode": "list",
"value": "[YOUR_ID]",
"cachedResultUrl": "",
"cachedResultName": ""
},
"table": {
"__rl": true,
"mode": "list",
"value": "[YOUR_ID]",
"cachedResultUrl": "",
"cachedResultName": ""
},
"columns": {
"value": {
"User": "[YOUR_NAME]",
"Memory": "={{ $fromAI('Memory') }}"
},
"schema": [
{
"id": "Memory",
"type": "string",
"display": true,
"removed": false,
"readOnly": false,
"required": false,
"displayName": "Memory",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "User",
"type": "string",
"display": true,
"removed": false,
"readOnly": false,
"required": false,
"displayName": "User",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "Created",
"type": "string",
"display": true,
"removed": true,
"readOnly": true,
"required": false,
"displayName": "Created",
"defaultMatch": false,
"canBeUsedToMatch": true
}
],
"mappingMode": "defineBelow",
"matchingColumns": [],
"attemptToConvertTypes": false,
"convertFieldsToString": false
},
"options": [],
"operation": "create"
},
"credentials": {
"airtableTokenApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 2.1
},
{
"id": "b8192282-2bd6-4a84-9ea2-0e0ac58673df",
"name": "Retrieve Memory Entries",
"type": "n8n-nodes-base.airtable",
"position": [
910,
175
],
"parameters": {
"base": {
"__rl": true,
"mode": "list",
"value": "[YOUR_ID]",
"cachedResultUrl": "",
"cachedResultName": ""
},
"sort": {
"property": [
{
"field": "Created"
}
]
},
"table": {
"__rl": true,
"mode": "list",
"value": "[YOUR_ID]",
"cachedResultUrl": "",
"cachedResultName": ""
},
"options": [],
"operation": "search",
"filterByFormula": "({User} = '[YOUR_NAME]')"
},
"credentials": {
"airtableTokenApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 2.1
},
{
"id": "70adb85f-d25b-41f1-b886-19ffe2303bb4",
"name": "Compile Memory List",
"type": "n8n-nodes-base.aggregate",
"position": [
685,
200
],
"parameters": {
"include": "specifiedFields",
"options": [],
"aggregate": "aggregateAllItemData",
"fieldsToInclude": "Memory, Created",
"destinationFieldName": "Memories"
},
"typeVersion": 1
},
{
"id": "c0a63e6d-1cc9-4c96-aa93-25e26031bcc3",
"name": "Combine Inputs",
"type": "n8n-nodes-base.merge",
"position": [
455,
275
],
"parameters": {
"mode": "combine",
"options": [],
"combineBy": "combineAll"
},
"typeVersion": 3.2
},
{
"id": "65fed632-8371-4b3a-a4a8-5b1441f4909b",
"name": "Telegram Incoming Hook",
"type": "n8n-nodes-base.telegramTrigger",
"position": [
1365,
285
],
"webhookId": "c44b1cc5-25c3-44ad-b0dd-689957f1bfd7",
"parameters": {
"updates": [
"message"
],
"additionalFields": {
"chatIds": "[YOUR_ID]",
"download": true
}
},
"credentials": {
"telegramApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1.2
},
{
"id": "0cbca25b-507e-4771-b2fe-9d316bbe7ded",
"name": "Send Telegram Reply",
"type": "n8n-nodes-base.telegram",
"position": [
0,
260
],
"webhookId": "80e65f83-b4db-4d48-9386-65bec5a6a380",
"parameters": {
"text": "={{ $json.output }}",
"chatId": "={{ $('Telegram Incoming Hook').item.json.message.chat.id }}",
"additionalFields": {
"appendAttribution": false
}
},
"credentials": {
"telegramApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1.2
},
{
"id": "9164db77-04c2-4f5e-a3a8-c641d2f823c6",
"name": "Web Search via SerpAPI",
"type": "@n8n/n8n-nodes-langchain.toolSerpApi",
"position": [
470,
540
],
"parameters": {
"options": []
},
"credentials": {
"serpApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1
},
{
"id": "a8ecc80e-1569-4df0-8f55-1b0d8a93346c",
"name": "Fetch Calendar Events",
"type": "n8n-nodes-base.googleCalendarTool",
"position": [
800,
235
],
"parameters": {
"options": [],
"timeMax": "={{ $now.plus({ week: 4 }) }}",
"calendar": {
"__rl": true,
"mode": "list",
"value": "",
"cachedResultName": ""
},
"operation": "getAll"
},
"credentials": {
"googleCalendarOAuth2Api": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1.3
},
{
"id": "fc24c542-e26a-4f8d-a7ae-3d108dd50d0c",
"name": "Create Calendar Event",
"type": "n8n-nodes-base.googleCalendarTool",
"position": [
1005,
215
],
"parameters": {
"end": "=",
"start": "=",
"calendar": {
"__rl": true,
"mode": "list",
"value": "",
"cachedResultName": ""
},
"additionalFields": []
},
"credentials": {
"googleCalendarOAuth2Api": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1.3
},
{
"id": "77665a24-7ca6-4d77-904f-21bbdbc2c3a1",
"name": "Modify Calendar Event",
"type": "n8n-nodes-base.googleCalendarTool",
"position": [
1205,
240
],
"parameters": {
"eventId": "={{ /*n8n-auto-generated-fromAI-override*/ $fromAI('Event_ID', ``, 'string') }}",
"calendar": {
"__rl": true,
"mode": "list",
"value": "",
"cachedResultName": ""
},
"operation": "update",
"updateFields": []
},
"credentials": {
"googleCalendarOAuth2Api": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1.3
},
{
"id": "f88dba24-8ab0-44c7-8cc3-438b15081fc2",
"name": "Remove Calendar Event",
"type": "n8n-nodes-base.googleCalendarTool",
"position": [
1405,
260
],
"parameters": {
"eventId": "={{ /*n8n-auto-generated-fromAI-override*/ $fromAI('Event_ID', ``, 'string') }}",
"options": [],
"calendar": {
"__rl": true,
"mode": "list",
"value": "",
"cachedResultName": ""
},
"operation": "delete"
},
"credentials": {
"googleCalendarOAuth2Api": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1.3
},
{
"id": "660ddae0-9950-4419-a459-822b846d86c1",
"name": "Retrieve Gmail Threads",
"type": "n8n-nodes-base.gmailTool",
"position": [
800,
585
],
"webhookId": "20f81dde-4f1b-4b08-a012-9248bcc2fed4",
"parameters": {
"simple": "={{ /*n8n-auto-generated-fromAI-override*/ $fromAI('Simplify', ``, 'boolean') }}",
"filters": [],
"options": [],
"operation": "getAll",
"returnAll": "={{ /*n8n-auto-generated-fromAI-override*/ $fromAI('Return_All', ``, 'boolean') }}"
},
"credentials": {
"gmailOAuth2": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 2.1
},
{
"id": "e6b4cf06-3546-4e51-9cc4-9ca2f977c8e6",
"name": "Fetch Gmail Message",
"type": "n8n-nodes-base.gmailTool",
"position": [
1005,
575
],
"webhookId": "724f7811-421a-4174-bb3e-abf670ccd5f2",
"parameters": {
"simple": "={{ /*n8n-auto-generated-fromAI-override*/ $fromAI('Simplify', ``, 'boolean') }}",
"options": [],
"messageId": "={{ /*n8n-auto-generated-fromAI-override*/ $fromAI('Message_ID', ``, 'string') }}",
"operation": "get"
},
"credentials": {
"gmailOAuth2": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 2.1
},
{
"id": "7c9ae477-0077-4084-98be-5265ac2941f6",
"name": "Grok Chat Model",
"type": "@n8n/n8n-nodes-langchain.lmChatXAiGrok",
"position": [
470,
155
],
"parameters": {
"model": "grok-4-0709",
"options": []
},
"credentials": {
"xAiApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1
},
{
"id": "ddaa9d4d-3ce1-4400-908d-1a059240a677",
"name": "Text or Voice Check",
"type": "n8n-nodes-base.if",
"position": [
1140,
270
],
"parameters": {
"options": [],
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "loose"
},
"combinator": "and",
"conditions": [
{
"id": "3c44273e-c862-447e-9512-afa1320ff950",
"operator": {
"type": "string",
"operation": "exists",
"singleValue": true
},
"leftValue": "={{ $json.message.text }}",
"rightValue": ""
}
]
},
"looseTypeValidation": true
},
"typeVersion": 2.2
},
{
"id": "c3464ae3-abe6-48b5-a37e-46384488a92f",
"name": "Transcribe Audio Clip",
"type": "@n8n/n8n-nodes-langchain.openAi",
"position": [
700,
630
],
"parameters": {
"options": [],
"resource": "audio",
"operation": "transcribe"
},
"credentials": {
"openAiApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1.8
},
{
"id": "5efdd1d7-3a22-49f2-a95e-3aa3349795ef",
"name": "Download Voice File",
"type": "n8n-nodes-base.telegram",
"position": [
930,
635
],
"webhookId": "e20e8498-b6aa-4264-b7d9-a6295c174360",
"parameters": {
"fileId": "={{ $json.message.voice.file_id }}",
"resource": "file",
"additionalFields": []
},
"credentials": {
"telegramApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 1.2
},
{
"id": "0dcc47a7-8c78-4328-bb78-d3b78cdd18c7",
"name": "Fetch Notion Database",
"type": "n8n-nodes-base.notionTool",
"position": [
1205,
590
],
"parameters": {
"simple": "={{ /*n8n-auto-generated-fromAI-override*/ $fromAI('Simplify', ``, 'boolean') }}",
"resource": "database",
"databaseId": {
"__rl": true,
"mode": "list",
"value": "",
"cachedResultUrl": "",
"cachedResultName": ""
}
},
"credentials": {
"notionApi": {
"id": "credential-id",
"name": ""
}
},
"typeVersion": 2.2
},
{
"id": "d5abb03f-5fd6-4103-b4f1-f357e448977a",
"name": "Conversation Orchestrator",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
225,
260
],
"parameters": {
"text": "={{ $json.text }}{{ $json.message.text }}",
"options": {
"systemMessage": "=# ROLE  \nYou are a friendly AI assistant.  \nYou are talking to ____. Your name is ____, your gender is ____.\n\n# RULES  \nWhen the user sends a new message, determine whether it contains information that should be saved in memory. Keep in mind that almost everything needs to be saved, as every detail is important to the user. Call the **Store Memory Record** tool to save the information. DO NOT inform the user that the information has been saved. Just continue to answer the question or perform the next tasks.  \nDo not use emojis, only in cases where it is truly appropriate.\n\n# Tools  \n## Store Memory Record  \nUse this tool to save information about the user. Extract and briefly summarize the details from the user's message and pass them to this tool.  \n## Web Search via SerpAPI  \nUse this tool if you need to find something on the Internet.  \n## Google Calendar  \nThese are several tools to be used if the user requests something related to the calendar and schedule.  \n## Gmail  \nThese are several tools to be used if the user requests something related to email and messages.  \n## Notion  \nThese are several tools to be used if the user requests something related to Notion, Neurocracy, or reading lists.\n\n# Memory  \nBelow are the latest significant memories collected about the user, including the date and time they were created.  \n!! IMPORTANT!  \nThink through your responses and consider the user's preferences!  \nAlso, keep in mind the date and time when the memory was saved to always have the most up-to-date information.\n\n{{ $json.Memories.toJsonString() }}"
},
"promptType": "define"
},
"typeVersion": 2.2
}
],
"active": false,
"pinData": [],
"settings": {
"executionOrder": "v1"
},
"versionId": null,
"connections": {
"Conversation Orchestrator": {
"main": [
[
{
"node": "Send Telegram Reply",
"type": "main",
"index": 0
}
]
]
},
"Text or Voice Check": {
"main": [
[
{
"node": "Combine Inputs",
"type": "main",
"index": 0
},
{
"node": "Retrieve Memory Entries",
"type": "main",
"index": 0
}
],
[
{
"node": "Download Voice File",
"type": "main",
"index": 0
}
]
]
},
"Combine Inputs": {
"main": [
[
{
"node": "Conversation Orchestrator",
"type": "main",
"index": 0
}
]
]
},
"Web Search via SerpAPI": {
"ai_tool": [
[
{
"node": "Conversation Orchestrator",
"type": "ai_tool",
"index": 0
}
]
]
},
"Compile Memory List": {
"main": [
[
{
"node": "Combine Inputs",
"type": "main",
"index": 1
}
]
]
},
"Download Voice File": {
"main": [
[
{
"node": "Transcribe Audio Clip",
"type": "main",
"index": 0
}
]
]
},
"Store Memory Record": {
"ai_tool": [
[
{
"node": "Conversation Orchestrator",
"type": "ai_tool",
"index": 0
}
]
]
},
"Retrieve Memory Entries": {
"main": [
[
{
"node": "Compile Memory List",
"type": "main",
"index": 0
}
]
]
},
"Brief Memory Buffer": {
"ai_memory": [
[
{
"node": "Conversation Orchestrator",
"type": "ai_memory",
"index": 0
}
]
]
},
"Telegram Incoming Hook": {
"main": [
[
{
"node": "Text or Voice Check",
"type": "main",
"index": 0
}
]
]
},
"Grok Chat Model": {
"ai_languageModel": [
[
{
"node": "Conversation Orchestrator",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Fetch Gmail Message": {
"ai_tool": [
[
{
"node": "Conversation Orchestrator",
"type": "ai_tool",
"index": 0
}
]
]
},
"Transcribe Audio Clip": {
"main": [
[
{
"node": "Combine Inputs",
"type": "main",
"index": 0
},
{
"node": "Retrieve Memory Entries",
"type": "main",
"index": 0
}
]
]
},
"Fetch Notion Database": {
"ai_tool": [
[
{
"node": "Conversation Orchestrator",
"type": "ai_tool",
"index": 0
}
]
]
},
"Retrieve Gmail Threads": {
"ai_tool": [
[
{
"node": "Conversation Orchestrator",
"type": "ai_tool",
"index": 0
}
]
]
},
"Create Calendar Event": {
"ai_tool": [
[
{
"node": "Conversation Orchestrator",
"type": "ai_tool",
"index": 0
}
]
]
},
"Remove Calendar Event": {
"ai_tool": [
[
{
"node": "Conversation Orchestrator",
"type": "ai_tool",
"index": 0
}
]
]
},
"Fetch Calendar Events": {
"ai_tool": [
[
{
"node": "Conversation Orchestrator",
"type": "ai_tool",
"index": 0
}
]
]
},
"Modify Calendar Event": {
"ai_tool": [
[
{
"node": "Conversation Orchestrator",
"type": "ai_tool",
"index": 0
}
]
]
}
}
}