Quick Start

Index a corpus and run your first query in under five minutes.

1. Create a project

The init command scaffolds a new RAG project from a template. The single positional argument is the template name (defaults to basic if omitted). Use --directory to set the target folder name:

rag-forge init basic --directory my-rag-app
cd my-rag-app

Available templates: basic, hybrid, agentic, enterprise, n8n. To scaffold a different one:

rag-forge init hybrid --directory my-rag-app

If you omit --directory, the project is created in a folder named after the template (e.g. ./basic).

2. Add some documents to index

The basic template doesn’t ship with sample documents — drop your own into a folder, or create a small example:

mkdir docs
echo "RAG-Forge is a CLI for building and evaluating RAG pipelines." > docs/example.md

3. Build the index

Point index at your source documents. The --source flag is required:

rag-forge index --source ./docs

This parses, chunks, embeds, and writes your documents to the configured vector store. By default it uses the mock embedding provider — set --embedding openai (or local) for real embeddings.

4. Query

rag-forge query "What does RAG-Forge do?"

5. Audit retrieval quality

rag-forge audit --golden-set eval/golden_set.json

This runs a baseline evaluation against the golden set in eval/golden_set.json and prints an RMM score (0–5).

What’s next

• Templates — start from a scaffold that matches your use case
• CLI Reference — every command and flag
• RAG Maturity Model — how RAG-Forge scores your pipeline