Overview

The CLI supports two ways to work. Workflow commands (ingest, query, chat) run the common path in one command. Stage commands (scrape, parse, fetch, chunk, embed, store, retrieve) create inspectable JSON files between stages.

Pipeline Map

Stage	CLI input	CLI output	Use it when
Extraction	URLs, files, folders, or APIs	JSON document files	You need source content in a common shape.
Chunking	Extraction JSON directory	`chunks.json`	You need retrieval-sized units.
Embedding	Chunk JSON directory	`embedded.json`	You need vectors for semantic search.
Storing	Embedded JSON directory	Vector database records	You need a searchable index.
Retrieval	Query string	Printed chunks and scores	You need search results without generation.
Chat	Query string or REPL	Answer, sources, optional history file	You need grounded answers.
Maintenance	Updated chunk JSON or chunk IDs	Replaced or removed vectors	You need to keep the index fresh.

Full Ingest Workflow

ragrails ingest \
  --docs files/handbook.pdf \
  --source-url https://example.com/docs \
  --markdown "# Refunds\n\nRefunds are available within 30 days." \
  --vector-db qdrant \
  --collection docs \
  --url http://localhost:6333 \
  --provider voyage \
  --model voyage-3

Manual Stage Pipeline

ragrails parse --files files/handbook.pdf --output-dir files/output/docs/
ragrails chunk --input-dir files/output/docs/ --output-dir files/output/chunks/
ragrails embed --input-dir files/output/chunks/ --output-dir files/output/embedded/ \
  --provider voyage --model voyage-3
ragrails store --input-dir files/output/embedded/ \
  --vector-db qdrant --collection docs --url http://localhost:6333

Query Pipeline

ragrails query "How do refunds work?" \
  --vector-db qdrant \
  --collection docs \
  --url http://localhost:6333 \
  --provider voyage \
  --model voyage-3 \
  --top-k 20 \
  --rerank \
  --rerank-top-k 5

Chat Pipeline

ragrails chat "How do refunds work?" \
  --vector-db qdrant \
  --collection docs \
  --url http://localhost:6333 \
  --llm-provider openai \
  --llm-model gpt-4o-mini \
  --history-file files/chat-history.json

Maintenance Loop

ragrails edit \
  --input-dir files/updated-chunks/ \
  --vector-db qdrant \
  --collection docs \
  --url http://localhost:6333

ragrails delete \
  --id old-policy-0 \
  --id old-policy-1 \
  --vector-db qdrant \
  --collection docs \
  --url http://localhost:6333

Choose the Entry Point

Need	Use
Build an index quickly	`ragrails ingest`
Search without generation	`ragrails query` or `ragrails retrieve`
Answer with sources	`ragrails chat "..."`
Interactive chat and tools	`ragrails chat`
Inspect intermediate artifacts	Stage commands with `--input-dir` / `--output-dir`
Replace or remove stale chunks	`ragrails edit` and `ragrails delete`

Extraction

Start with source inputs.

Maintenance

Keep stored chunks fresh.

Get Started

Workflows

Pipeline

Maintenance

Pipeline Map

Full Ingest Workflow

Manual Stage Pipeline

Query Pipeline

Chat Pipeline

Maintenance Loop

Choose the Entry Point

Extraction

Maintenance

​Pipeline Map

​Full Ingest Workflow

​Manual Stage Pipeline

​Query Pipeline

​Chat Pipeline

​Maintenance Loop

​Choose the Entry Point

Extraction

Maintenance

Pipeline Map

Full Ingest Workflow

Manual Stage Pipeline

Query Pipeline

Chat Pipeline

Maintenance Loop

Choose the Entry Point