Mindloom turns a pile of technical articles, documentation, and tutorials into a structured, searchable knowledge base. Paste in any technical text and Mindloom classifies it into one of eight categories, extracts its keywords, finds semantically related content already in your corpus, and makes everything discoverable by meaning — not just by matching words.Documentation Index
Fetch the complete documentation index at: https://mintlify.com/No-Country-simulation/G9-LATAM-Team-58/llms.txt
Use this file to discover all available pages before exploring further.
Introduction
Learn what Mindloom does, how the system works end-to-end, and which problem it solves.
Quickstart
Run Mindloom locally in minutes and ingest your first piece of technical content.
API Reference
Explore every endpoint: ingest content, search by meaning, browse the knowledge map.
Architecture
Understand the four-layer system: Web, API, Inference Service, and Oracle ADB.
What Mindloom does
Semantic Search
Find content by meaning using multilingual sentence embeddings and Oracle Vector Search.
Auto-Classification
Every ingested item is assigned to one of 8 categories — Backend, Frontend, Databases, and more.
Knowledge Map
Visualize the entire corpus as a 2D scatter plot, colored by category, powered by UMAP.
Batch Upload
Ingest hundreds of items at once by uploading a CSV file via the web UI or API.
Get up and running
Clone and configure
Clone the repository and copy
.env.example to .env, filling in your Oracle Autonomous Database credentials.Start the services
Run
docker compose up to bring up the API and inference containers. The inference service downloads the model artifact from OCI Object Storage on first startup.Launch the web UI
In the
web/ directory run npm run dev to start the React development server at http://localhost:5173.Mindloom requires Java 25, Python 3.12, Node 22, and Docker. Each service documents its own environment variables and configuration in its own
README.md.