@unwarkz/n8n-nodes-assistant-set

Full AI-assistant node set for n8n — Mem0 intelligent memory, Gotenberg PDF conversion, Telegram Bot messaging, Outline Wiki knowledge base, and Qdrant vector store. All nodes include both regular workflow variants and AI Agent tool variants.

Please star the repo if you use my nodes - I will know that it is needed to someone except me.

Each component is also available as a standalone package:

• @unwarkz/n8n-nodes-mem0
• @unwarkz/n8n-nodes-gotenberg-pdf
• @unwarkz/n8n-nodes-telegram-bot
• @unwarkz/n8n-nodes-outline-wiki
• @unwarkz/n8n-nodes-qdrant

Installation

Via Community Nodes panel (recommended)

1. Go to Settings > Community Nodes and click Install.
2. Enter @unwarkz/n8n-nodes-assistant-set.
3. Confirm and restart n8n.

Manual installation

cd ~/.n8n/nodes
npm install @unwarkz/n8n-nodes-assistant-set

Restart n8n to load the nodes.

Credential Setup

Included Nodes

Mem0 — Intelligent Memory Layer

Mem0 Memory — Context Retrieval Modes

The memory node implements loadMemoryVariables to inject context into the agent and saveContext to persist conversation turns back to Mem0 automatically.

Gotenberg — PDF Conversion

Requires a running Gotenberg instance:

docker run --rm -p 3000:3000 gotenberg/gotenberg:8

Gotenberg Operations

Telegram Bot — Full Bot API

Telegram Bot AI Tools

Telegram Bot Node — Full API Coverage

• Messages: sendMessage, forwardMessage, copyMessage, sendPhoto, sendAudio, sendDocument, sendVideo, sendAnimation, sendVoice, sendVideoNote, sendMediaGroup, sendLocation, sendVenue, sendContact, sendPoll, sendDice, sendChatAction, editMessageText, editMessageCaption, editMessageMedia, editMessageReplyMarkup, deleteMessage, deleteMessages
• Chat Management: getChat, getChatAdministrators, getChatMemberCount, getChatMember, banChatMember, unbanChatMember, restrictChatMember, promoteChatMember, setChatPermissions, exportChatInviteLink, createChatInviteLink, editChatInviteLink, revokeChatInviteLink, approveChatJoinRequest, declineChatJoinRequest, setChatPhoto, setChatTitle, pinChatMessage, unpinChatMessage, leaveChat
• Bot Info: getMe, logOut, getMyCommands, setMyCommands, deleteMyCommands, getMyName, setMyName, getMyDescription, setMyDescription
• Webhooks: setWebhook, deleteWebhook, getWebhookInfo
• Files: getFile (download with binary data)
• Stickers: sendSticker, getStickerSet, createNewStickerSet, addStickerToSet, deleteStickerFromSet
• Inline: answerInlineQuery, answerWebAppQuery
• Payments: sendInvoice, createInvoiceLink, answerShippingQuery, answerPreCheckoutQuery
• Forum Topics: createForumTopic, editForumTopic, closeForumTopic, reopenForumTopic, deleteForumTopic
• Games: sendGame, setGameScore, getGameHighScores
• Updates: getUpdates

Outline Wiki — Knowledge Base

Outline Wiki — Supported Resources and Operations

Attachments use a two-step upload: POST /api/attachments.create to obtain a signed S3 URL, then a multipart POST directly to S3.

Qdrant — Vector Store Knowledgebase

Qdrant AI Tools — Operations

The Qdrant Store node supports multimodal embedders with automatic base64 fallback for binary files.

Cross-Tool Binary File Interoperability

AI Tools nodes exchange binary files via n8n native binary references — no base64 encoding in the AI context window. Tools that produce files store them via prepareBinaryData and return a binaryPropertyName. Tools that consume files accept a binary_property_name parameter and read data via getBinaryDataBuffer. Compatible with N8N_DEFAULT_BINARY_DATA_MODE=filesystem and database.

Example AI Agent workflows:

Download file from Telegram   →  Convert with Gotenberg         →  Send back via Telegram
Generate PDF from URL         →  Store reference in Mem0        →  Send via Telegram
Receive Telegram attachment   →  Import into Outline Wiki       →  Embed as Outline attachment
Export Outline document       →  Convert to PDF with Gotenberg  →  Send PDF via Telegram
Download image from Telegram  →  Embed with Qdrant Store        →  Search similar images

Node Reference

Mem0 CRUD Node — Parameters

Common Parameters

Memory — Advanced Search Filters

Build structured filter rules in the UI:

• Field: e.g., memory, user_id, metadata.category
• Operation: Equals, Not Equals, Contains, Greater Than, Less Than
• Value: string; numbers, booleans, and JSON are parsed automatically
• Rules are combined with AND and sent to the v2 search endpoint

Mem0 Memory — Advanced Options

Self-Hosted Mem0 API Compatibility

Additional self-hosted-only endpoints accessible via the Mem0 node:

• GET /health — service status
• GET /config — current LLM/embedder configuration
• POST /config/switch — switch LLM provider
• POST /reset — reset all memories

Usage Examples

1. Search memories with advanced filters

1. Add a Mem0 node. Set Resource = Memory, Operation = Advanced Search.
2. Enter a query (e.g., "ACME client preferences").
3. Add a filter rule: Field = metadata.segment, Operation = Equals, Value = enterprise.
4. Adjust Top K, Fields, and Rerank as needed.
5. Connect the output to your workflow.

2. Connect Mem0 Memory to an AI Agent (Hybrid mode)

1. Add a Mem0 Memory node and configure credentials.
2. Set Context Retrieval Mode to Hybrid.
3. Set Query to ={{ $json.lastUserMessage }} and Memory Key to chat_history.
4. Adjust Last N, Top K, Alpha, and Half-life as needed.
5. Connect the ai_memory output to the Memory port of the AI Agent node.

3. Convert a URL to PDF with AI Agent

1. Add a Gotenberg AI Tools node. Select the gotenberg_url_to_pdf tool.
2. Connect the ai_tool output to the Tools port of the AI Agent node.
3. The agent can now generate PDFs from URLs on demand.

4. Give the AI Agent Telegram messaging tools

1. Add a Telegram Bot AI Tools node. Configure credentials.
2. Enable the tools you need (e.g., telegram_send_message, telegram_get_file).
3. Connect the ai_tool output to the Tools port of the AI Agent node.

5. Embed documents into Qdrant

1. Add a Qdrant Store AI Tools node. Configure credentials.
2. Connect an Embedding model sub-node to the Embedding input.
3. Connect the ai_tool output to the Tools port of the AI Agent node.
4. The agent can now embed and retrieve documents from Qdrant.

Requirements

• n8n >= 1.0.0
• For Gotenberg nodes: a running Gotenberg instance
• For Mem0 nodes: a Mem0 Cloud account or self-hosted Mem0 API
• For Telegram nodes: a Telegram Bot token from @BotFather
• For Outline nodes: a running Outline instance + API token
• For Qdrant nodes: a running Qdrant instance

License

MIT. Do not commit sensitive information (API keys, personal data) to your workflows or to the npm registry.