CLI Reference - Ragrails

This page is generated from the Click command tree. Run RAGRAILS_SOURCE=/path/to/ragrails uv run python scripts/generate-reference.py from the docs repo after CLI changes.

Commands read defaults from .ragrails.toml; flags override per run. The setup wizard is launched by running bare ragrails with no subcommand; there is no ragrails setup command.

Setup

Command	Description
`ragrails`	Interactive setup wizard (writes `.ragrails.toml`)

The setup wizard prompts for vector store, embedding, LLM, and reranker defaults. If .ragrails.toml already exists, bare ragrails shows the current config and lets you edit, reset, or exit.

Wizard section	Prompts
Vector store	Provider, collection, URL
Embedding	Provider, model
LLM	Provider, model, max tokens
Reranking	Enable flag, provider, model
Advanced defaults	Chunking, batch sizes, retrieval limits, query rewrite, intent routing, history compaction

Commands

Command	Description
`ragrails chat`	Run one SDK chat turn, or start the interactive chat CLI with no query.
`ragrails chunk`	Split extraction JSON into chunks.
`ragrails delete`	Delete stored chunks by exact chunk ID.
`ragrails doctor`	Inspect `.ragrails.toml`, packages, environment variables, and optional vector DB connectivity.
`ragrails edit`	Re-embed and replace stored chunks by exact chunk ID.
`ragrails embed`	Embed chunk JSON files.
`ragrails fetch`	Fetch a REST API endpoint into extraction JSON.
`ragrails ingest`	Run extraction, chunking, embedding, and storage in one command.
`ragrails parse`	Parse local files, folders, or file URLs into extraction JSON.
`ragrails query`	Run query embedding and vector retrieval through the SDK pipeline.
`ragrails retrieve`	Retrieve chunks relevant to a query.
`ragrails scrape`	Scrape one or more URLs into extraction JSON.
`ragrails setup-url`	Install the Playwright browser runtime required for URL extraction.
`ragrails store`	Store embedded chunk JSON files in a vector database.