Alejandro Scuncia
Support & Services Leader exploring GenAI and automation. Currently leading a global support team while building and learning through real-world AI workflows — from smarter knowledge agents to complex integrations. Here to explore, share, and connect with the n8n community.
Templates by Alejandro Scuncia
Smarter RAG agents with enriched retrieval and modular workflows
An extendable RAG template to build powerful, explainable AI assistants — with query understanding, semantic metadata, and support for free-tier tools like Gemini, Gemma and Supabase. Description This workflow helps you build smart, production-ready RAG agents that go far beyond basic document Q&A. It includes: ✅ File ingestion and chunking ✅ Asynchronous LLM-powered enrichment ✅ Filterable metadata-based search ✅ Gemma-based query understanding and generation ✅ Cohere re-ranking ✅ Memory persistence via Postgres Everything is modular, low-cost, and designed to run even with free-tier LLMs and vector databases. Whether you want to build a chatbot, internal knowledge assistant, documentation search engine, or a filtered content explorer — this is your foundation. ⚙️ How It Works This workflow is divided into 3 pipelines: 📥 Ingestion Upload a PDF via form Extract text and chunk it for embedding Store in Supabase vector store using Google Gemini embeddings 🧠 Enrichment (Async) Scheduled task fetches new chunks Each chunk is enriched with LLM metadata (topics, use_case, risks, audience level, summary, etc.) Metadata is added to the vector DB for improved retrieval and filtering 🤖 Agent Chat A user question triggers the RAG agent Query Builder transforms it into keywords and filters Vector DB is queried and reranked The final answer is generated using only retrieved evidence, with references Chat memory is managed via Postgres 🌟 Key Features Asynchronous enrichment → Save tokens, batch process with free-tier LLMs like Gemma Metadata-aware → Improved filtering and reranking Explainable answers → Agent cites sources and sections Chat memory → Persistent context with Postgres Modular design → Swap LLMs, rerankers, vector DBs, and even enrichment schema Free to run → Built with Gemini, Gemma, Cohere, Supabase (free tier-compatible) 🔐 Required Credentials |Tool|Use| |-|-|-| |Supabase w/ PostreSQL|Vector DB + storage| |Google Gemini/Gemma|Embeddings & LLM| |Cohere API|Re-ranking| |PostgreSQL|Chat memory| 🧰 Customization Tips Swap extractFromFile with Notion/Google Drive integrations Extend Metadata Obtention prompt to fit your domain (e.g., financial, legal) Replace LLMs with OpenAI, Mistral, or Ollama Replace Postgre Chat Memory with Simple Memory or any other Use a webhook instead of a form to automate ingestion Connect to Telegram/Slack UI with a few extra nodes 💡 Use Cases Company knowledge base bot (internal docs, SOPs) Educational assistant with smart filtering (by topic or level) Legal or policy assistant that cites source sections Product documentation Q&A with multi-language support Training material assistant that highlights risks/examples Content Generation 🧠 Who It’s For Indie developers building smart chatbots AI consultants prototyping Q&A assistants Teams looking for an internal knowledge agent Anyone building affordable, explainable AI tools 🚀 Try It Out! Deploy a modular RAG assistant using n8n, Supabase, and Gemini — fully customizable and almost free to run. 📁 Prepare Your PDFs Use any internal documents, manuals, or reports in PDF format. Optional: Add Google Drive integration to automate ingestion. 🧩 Set Up Supabase Create a free Supabase project Use the table creation queries included in the workflow to set up your schema. Add your supabaseUrl and supabaseKey in your n8n credentials. > 💡 Pro Tip: Make sure you match the embedding dimensions to your model. This workflow uses Gemini text-embedding-04 (768-dim) — if switching to OpenAI, change your table vector size to 1536. 🧠 Connect Gemini & Gemma Use Gemini/Gemma for embeddings and optional metadata enrichment. Or deploy locally for lightweight async LLM processing (via Ollama/HuggingFace). ⚙️ Import the Workflow in n8n Open n8n (self-hosted or cloud). Import the workflow file and paste your credentials. You’re ready to ingest, enrich, and query your document base. 💬 Have Feedback or Ideas? I’d Love to Hear This project is open, modular, and evolving — just like great workflows should be :). If you’ve tried it, built on top of it, or have suggestions for improvement, I’d genuinely love to hear from you. Let’s share ideas, collaborate, or just connect as part of the n8n builder community. 📧 ascuncia.es@gmail.com 🔗 Linkedin
By Alejandro ScunciaTicket triage for Jira Service Management with Gemini AI audit and guidance
An extendable triage workflow that classifies severity, sets components, and posts actionable guidance for support engineers using n8n + Gemini + Cache Augmented Generation (CAG). Designed for Jira Service Management, but easily adaptable to Zendesk, Freshdesk, or ServiceNow. --- Description Support teams loose valuable time when tickets are misclassified: wrong severity, missing components, unclear scope. Engineers end up re-routing issues and chasing missing info instead of solving real problems. This workflow automates triage by combining domain rules with AI-driven classification and guidance, so engineers receive better-prepared tickets. It includes: ✅ Real-time ticket capture via webhook ✅ AI triage for severity and component ✅ CAG-powered guidance: 3 next steps + missing info ✅ Internal audit comment with justifications & confidence ✅ Structured metrics for reporting --- ⚙️ How It Works This workflow runs in 4 stages: 📥 Entry & Setup Webhook triggers on ticket creation Loads domain rules (priority policy, components, guidance templates) Sets confidence threshold & triage label 🧠 AI Analysis (Gemini + CAG) Builds structured payload with ticket + domain context Gemini proposes severity, component, guidance, missing info Output normalized for safe automation (valid JSON, conservative confidence) 🤖 Update & Audit Updates fields (priority, component, labels) if confidence ≥ threshold Posts internal audit comment with: 3 next steps Missing info to request Justifications + confidence 📊 Metrics Captures applied changes, confidence scores, and API statuses Enables reliability tracking & continuous improvement --- 🌟 Key Features CAG-powered guidance → lightning-fast, context-rich next steps Explainable automation → transparent audit comments for every decision Domain-driven rules → adaptable to any product or support domain Portable → swap JSM with Zendesk, Freshdesk, ServiceNow via HTTP nodes --- 🔐 Required Credentials | Tool | Use | |------|-----| | Jira Service Management | Ticketing system (API + comments) | | Google Gemini/Gemma | LLM analysis | | HTTP Basic Auth | For Jira API requests (bot user) | ⚠️ Setup tip: create a dedicated bot user in Jira Service Management with an API token. This ensures clean audit logs, proper permissions, and avoids mixing automation with human accounts. --- 🧰 Customization Tips Replace https://your-jsm-url/... with your own Jira Service Management domain. Update the credentials with the bot user’s API token created above. Swap Jira Service Management nodes with other ticketing systems like Zendesk, Freshdesk, or ServiceNow. Extend the domain schema (keywords, guidance_addons) to fit your product or support environment. --- 🗂️ Domain Schema This workflow uses a domain-driven schema to guide triage. It defines: Components → valid areas for classification Priority policies & rules → how severity is determined Keywords → domain-specific signals (e.g., “API error”, “all users affected”) Guidance addons → contextual next steps for engineers No-workaround phrases → escalate severity if present ✨ The full domain JSON (with complete keyword & guidance mapping) is included as a sticky note inside the workflow. --- 💡 Use Cases Automated triage for IT & support tickets Incident classification with outage/security detection Contextual guidance for engineers in customer support Faster escalation and routing of critical issues --- 🧠 Who It’s For Support teams running Jira Service Management Platform teams automating internal ticket ops AI consultants prototyping practical triage workflows Builders exploring CAG today, RAG tomorrow --- 🚀 Try It Out! ⚙️ Import the Workflow in n8n (cloud or self-hosted). 🔑 Add Credentials (JSM API + Gemini key). ⚡ Configure Setup (confidence threshold, triage label, domain rules). 🔗 Connect Webhook in JSM → issue_created → n8n webhook URL. 🧪 Test with a Ticket → see auto-updates + AI audit comment. 🔄 Swap the Ticketing System → adapt HTTP nodes for Zendesk, Freshdesk, or ServiceNow. --- 💬 Have Feedback or Ideas? I’d Love to Hear This project is open, modular, and evolving. If you try it, adapt it, or extend it, I’d love to hear your feedback — let’s improve it together in the n8n builder community. 📧 ascuncia.es@gmail.com 🔗Linkedin
By Alejandro ScunciaManage Trello tasks with AI assistants via MCP server
A lean MCP Server that exposes the essential Trello tools for everyday task management. Built for clean, reliable LLM automation with n8n, ChatGPT, or Gemini. Description A minimal MCP Server offering the core Trello operations—create, update, search, list, and comment—designed to keep task management simple and predictable. Each tool has a clear, LLM-friendly schema, ensuring safe automation without heavy context or complex payloads. Use it to create tasks, move cards between lists, update details, add comments, or retrieve board structure. All through natural language. --- ⭐ Key Features Lean Toolset – only the operations needed for real task flows LLM-Ready – minimal, explicit parameters for safe execution Extendable – add more Trello endpoints or custom logic easily Predictable Updates – partial updates with safe defaults MCP + Tool Calling – works in n8n Agents, ChatGPT, Gemini, and MCP clients --- 🧠 Examples (Natural Language) “What tasks do I still have pending from last week?” → Searches by due date and status for weekly planning. “Move all tasks due yesterday into today’s list and add a ‘rescheduled’ comment.” → Combines search + update + comment. “Create a new card called ‘Prepare onboarding materials’ for next Monday.” → Simple card creation with due date handling. --- ⚙️ How It Works This MCP Server exposes 6 core tools: 📥 Create Card Create tasks with title, description, and optional due date. 🔍 Search Cards Find cards using Trello’s native search syntax (keywords, lists, due filters). 🗂 List Backlog & List All Lists Provide structure and situational awareness for agents. ✏️ Update Card Rename, update description, adjust due dates, or move cards. 💬 Add Comment Add quick notes to any card without modifying other fields. --- 🔐 Required Credentials | Service | Use | |--------|-----| | Trello API | API Key + Token for card, list, search, and comment operations | Use a dedicated Trello automation token for clean audit logs. Please check Authorizing With Trello's REST API --- 🧰 Customization Tips Replace board/list IDs with your own Add labels, checklists, or custom fields if needed Add domain logic (priority rules, conventions, or workflows) Use listalllists in agents to give LLMs workflow awareness --- 🧠 Who It’s For Automation builders exploring MCP AI assistants for personal task planning Teams with simple Trello workflows Consultants building lightweight agents --- 🚀 Try It Out 1) Import the MCP workflow into n8n 2) Add your Trello API credentials 3) Replace IDs with your board/list values 4) Test with ChatGPT or Gemini via natural-language prompts 5) (Optional) Register it as an MCP Server in VS Code or any MCP client --- 💬 Feedback Welcome This MCP Server is intentionally simple and extendable. If you adapt or evolve it, I’d love to hear your ideas. 📧 ascuncia.es@gmail.com 🔗 Linkedin
By Alejandro Scuncia