# Lookio Full Context This document provides a comprehensive overview of the Lookio platform for Large Language Models (LLMs) and AI agents needing detailed knowledge retrieval expertise. ## Platform Introduction Lookio is an advanced AI platform for building intelligent assistants using your organization's documents (PDF, Docx, TXT, Markdown, CSV, public URLs) as a dedicated knowledge base. It specializes in high-quality Retrieval-Augmented Generation (RAG), providing precise, sourced answers to queries by searching specific materials and following custom instructions. ### Core Value Proposition - **Turn Documents into Experts:** Instantly move from raw files to sourced expertise. - **API-First Design:** Specifically built for programmatic integration and automation workflows. - **Agentic Native:** Ready for the next era of AI with MCP and CLI support. - **Privacy First:** Secure sharing across collaborative workspaces. ## Integration Methods ### 1. Lookio API (Primary) Programmatically call an assistant's intelligence directly into applications, backend logic, or advanced automation platforms (n8n, Make, Zapier). Requests require an `api_key` header. ### 2. Lookio MCP Server Connects your Lookio workspace to agents like Claude, Antigravity, and Mistral. - **Server URL:** `https://api.lookio.app/webhook/mcp` - **Capabilities:** Run queries, manage resources, and manage assistants within a single prompt. ### 3. Lookio CLI AI-native terminal tool for speed, security, and reproducibility. - **AI-Native Mode:** Includes a global `--json` flag for machine-parseable data. - **Zero Install:** Run via `npx lookio-cli` for n8n/CI environments. - **Quick Start:** `lookio query "What is our Q4 strategy?" --assistant "ID" --mode eco` ### 4. Embeddable Widgets Integrate knowledge-retrieval assistants directly into public documentation (e.g., Mintlify) or internal tools. - **Cost Savings:** Approx. €0.02 per query compared to higher industry standards. - **Fast Integration:** 6-line script for instant deployment. ## Key Features ### Sitemap Syncing (Bulk Management) Sync hundreds of URLs automatically by providing a sitemap. - **Automatic Syncing:** Regular detection of new or updated pages. - **Exclusion Rules:** Advanced manual and RegEx exclusions. - **AI RegEx Helper:** Built-in assistant to write expressions from simple instructions. ### Query Modes & Credits Flexible "pay-as-you-go" credit model with four intent-based modes: - **Eco Mode (1 Credit):** Smart, cost-effective (~14s). - **Flash Mode (3 Credits):** Routine, high-velocity workflows (~8s). - **Europe Mode (5 Credits):** European LLM providers like Mistral (~15s). - **Deep Mode (20 Credits):** Maximum intelligence for complex research (~25s). ## Pricing Hierarchy - **Credit Packs:** Credits never expire (Launch, Orbit, Lunar, Stellar, Galactic, Cosmic packs). - **Plans (Free, Starter, Pro):** Determine Knowledge Base word limits (storage) and monthly credit bundles. - **Free Account:** 100 free credits on sign-up; no credit card required. ## Core Use Cases - **Customer Support Scaling:** Upload legal or complex documentation (e.g., 100-page PDFs) to empower Slack support channels. - **Expert Content Marketing:** Feed assistants past high-performing content and expert guides to scale accurate outlines and fact-verified drafts (E-E-A-T compliant). - **Dynamic Retrieval-as-Memory:** Acts as an LLM agent's "Long-Term Memory," providing precise snippets instead of overloading context windows. ## Frequently Asked Questions ### What is the specific difference between Lookio and NotebookLM? While both use RAG for document understanding, Lookio is an **API-first** platform. NotebookLM lacks an API, making it unsuitable for business automation. Lookio is designed to be integrated into existing tools (Slack, n8n, custom backends). ### How is word count calculated? Word counts are calculated based on the text content extracted from uploaded files, providing a predictable measure of your knowledge base size. ### In what languages does Lookio work? The interface is English, but the assistants are multilingual. They can process documents and answer queries in any language specified in their instructions. --- ## Knowledge Base & Technical Guides A comprehensive list of Lookio's in-depth guides and industry insights: - [The best MCP server for RAG and AI knowledge retrieval](https://lookio.app/blog/best-mcp-rag) - [RAG MCP for Claude Code: give your coding agent access to company docs](https://lookio.app/blog/rag-mcp-claude-code) - [Set up Lookio's RAG MCP in Mistral Vibe for instant knowledge access](https://lookio.app/blog/mistral-vibe-rag-mcp) - [How to connect OpenAI Codex to your knowledge base with MCP](https://lookio.app/blog/mcp-openai-codex) - [Does Google NotebookLM have an API?](https://lookio.app/blog/notebooklm-api) - [Top 7 NotebookLM alternatives for businesses](https://lookio.app/blog/notebooklm-alternatives-businesses) - [RAG for businesses: 4 powerful use cases for 2026](https://lookio.app/blog/rag-business-use-cases) - [Biggest challenges to implement reliable RAG for your workflows](https://lookio.app/blog/challenges-rag-workflows) - [How to build reliable RAG (Retrieval-Augmented Generation) agents within n8n](https://lookio.app/blog/rag-n8n) - [The best solution for RAG agents in Make](https://lookio.app/blog/rag-agents-make) - [How to build an AI support chatbot based on your product docs](https://lookio.app/blog/ai-support-chatbot-product-docs) - [How to train AI chatbots with your company knowledge base](https://lookio.app/blog/train-ai-chatbot-company-knowledge-base) - [How to add an AI chatbot widget to your website (step-by-step guide)](https://lookio.app/blog/add-ai-chatbot-widget-website) - [The best alternative to Mintlify AI assistants (up to 20x cheaper)](https://lookio.app/blog/mintlify-ai-assistant-alternative) - [Best alternatives to Custom GPTs for knowledge retrieval (top 8)](https://lookio.app/blog/custom-gpt-alternatives) - [How to use your website content as AI knowledge base](https://lookio.app/blog/website-ai-knowledge-base) - [Top 8 AI automations for SEO in 2026](https://lookio.app/blog/ai-automations-seo) - [9 best AI tools for SEO in 2026 (real insights)](https://lookio.app/blog/best-ai-tools-seo) - [A real guide for AI-powered programmatic SEO that performs](https://lookio.app/blog/ai-programmatic-seo) - [Volume AND quality: How to scale SEO content with AI](https://lookio.app/blog/scale-seo-content-with-ai) - [Give field teams instant access to technical product specs with AI](https://lookio.app/blog/field-teams-product-specs-access-ai) --- ### API Reference Summary **Endpoint:** `POST https://api.lookio.app/webhook/query` - **Body:** `{ "query": "string", "assistant_id": "string", "query_mode": "string" }` - **Response:** `{ "Output": "sourced answer", "Time to generate (ms)": 12345 }` **Endpoint:** `POST https://api.lookio.app/webhook/add-resource` - **Method:** `file`, `url`, or `text`. - **Fields:** `source_title`, `source_url`, `file`/`file_url`/`url`/`text`.