---
title: Non-Goals
sidebar_position: 3
---
TeaRAGs is focused on a specific problem space. Understanding what it does **not** aim to do is as important as understanding what it does.
## Not a General-Purpose Vector Database
TeaRAGs uses Qdrant as its storage backend, but it is not a general-purpose vector database interface. It is purpose-built for code search with development history awareness.
## Not *Another* Code Analysis Tool
TeaRAGs **can** analyze code — and does it well. Trajectory enrichment provides real metrics at **function-level granularity**: stability, churn, ownership, bug-fix rates, code age — per function, per method, not just per file. This makes it a powerful tool for hotspot detection, tech debt scoring, and ownership mapping.
However, the **primary focus** is intelligent code generation, not analysis for its own sake. The analysis capabilities exist to make retrieval smarter — so agents find the *right* code to learn from, not just similar code. TeaRAGs is not a static analysis tool, linter, or code quality dashboard. It's an intelligence layer that *uses* analysis signals to produce better search results and better-informed code generation.
## Not a Replacement for grep/ripgrep
For exact string matching, use ripgrep. TeaRAGs excels at semantic queries ("how does authentication work?") not literal text search ("find all occurrences of `AUTH_TOKEN`").
## Not a CI/CD Component
TeaRAGs is designed for interactive use by developers and AI agents. It is not optimized for pipeline automation or batch processing in CI/CD workflows.
## Not a Dinosaur Simulator
Despite the logo, TeaRAGs will not help you clone dinosaurs, brew tea, or type with tiny arms. It *will* help your coding agent make smarter decisions — which is arguably more useful. 🦖
