Getting Started with Terraphim AI
Welcome to Terraphim AI! This guide will help you get up and running quickly, whether you're a developer, researcher, or end user.
π Quick Start (5 Minutes)
Choose Your Installation Method
Option 1: Rust CLI/TUI (Recommended for Power Users)
Option 2: Node.js Package (Great for Web Integration)
# or with Bun
Option 3: Python Library (Perfect for Data Processing)
Option 4: Desktop Application (GUI Users)
Download from GitHub Releases
Your First Query
After installation, try your first semantic search:
# Rust CLI
# Node.js
;
;
# Python
π― What Can Terraphim AI Do?
Core Capabilities
- π Semantic Search: Find information across multiple data sources
- π§ Knowledge Graph: Understand relationships between concepts
- π¬ AI Chat: Interactive conversations with your data
- π Smart Linking: Automatic markdown/html/wiki linking
- π Auto-Update: Always up-to-date with latest features
Data Sources You Can Search
- Local Files: Your markdown, text, and code files
- GitHub Repositories: Search across 500,000+ public repositories
- Team Knowledge: Jira, Confluence, SharePoint integration
- Stack Overflow: Programming questions and answers
ποΈ Architecture Overview
Terraphim AI uses a modular, privacy-first architecture:
βββββββββββββββββββ ββββββββββββββββββββ βββββββββββββββββββ
β Rust Core β β Node.js Bindingsβ β Python Bindings β
β (terraphim_ β β (@terraphim/ β β (terraphim- β
β service) β β autocomplete) β β automata) β
βββββββββββ¬ββββββββ βββββββββββ¬ββββββββββ βββββββββββ¬ββββββββββ
β β β
βββββββββββββββββββββββββββ΄βββββββββββββββββββββ
βββββββββββββββββββββββββββ
β Knowledge Graph β
β (terraphim_rolegraph) β
β (terraphim_automata) β
βββββββββββββββββββββββββββπ§ Configuration Basics
Environment Setup
1. Basic Configuration
Create a configuration file at ~/.config/terraphim/config.toml:
[data]
default_data_path = "~/Documents/terraphim"
index_documents = true
[search]
default_scorer = "tfidf" # Options: bm25, tfidf, jaccard
max_results = 20
[llm]
provider = "ollama" # Options: ollama, openrouter, claude
model = "llama3.2:3b"
temperature = 0.72. Data Source Configuration
Add your data sources:
[sources.local_files]
path = "~/Documents"
file_types = ["md", "txt", "rst"]
[sources.github]
enabled = true
languages = ["rust", "python", "javascript"]π Next Steps
For Different User Types
π¬ Researchers & Data Scientists
π Web Developers
π₯βπΌ Business Users
π§ System Administrators
β Need Help?
Common Questions
Q: Which package should I use? A:
- Developers: Rust CLI (
terraphim-agent) - Web Integration: Node.js (
@terraphim/autocomplete) - Data Analysis: Python (
terraphim-automata) - GUI Users: Desktop application
Q: How do I add my own data? A: See Configuration Guide for detailed instructions
Q: Can I use Terraphim offline? A: Yes! Terraphim is privacy-first and works completely offline with embedded defaults
Getting Support
- π Documentation: Full Documentation
- π¬ Discord: Join our Community
- ποΈ Discourse: Discussion Forum
- π Issues: Report on GitHub
π You're Ready!
You now have Terraphim AI installed and configured. Start exploring your knowledge graph with semantic search powered by AI assistance.
Try these commands first:
Welcome to the future of privacy-first AI assistance! π
Last Updated: December 20, 2025 Part of: Terraphim AI Documentation v1.3.0