REST API Overview - Ragrails

The Ragrails REST API exposes the same pipeline as the SDK through HTTP. Use it when your application is not Python, when you want a service boundary around indexing and chat, or when clients need to upload documents directly.

pip install "ragrails[server-qdrant]"
ragrails-api

The server starts on http://127.0.0.1:8000.

Tool	URL
Swagger UI	`http://127.0.0.1:8000/docs`
ReDoc	`http://127.0.0.1:8000/redoc`
OpenAPI schema	`http://127.0.0.1:8000/v1/openapi.json`
Health check	`GET /v1/health`

Workflows

Workflow	Endpoint	Use it for
Ingest	`POST /v1/pipelines/ingest`	Extract, chunk, embed, and store content in one request.
Query	`POST /v1/pipelines/query`	Embed a search query and retrieve matching chunks.
Chat	`POST /v1/chat`	Retrieve context and generate a grounded answer.

Pipeline

Start with the pipeline overview, or call individual stages when you want to inspect or persist each intermediate JSON result.

Stage	Endpoints	Details
Extraction	`/v1/ingest/url`, `/v1/ingest/docs`, `/v1/ingest/api`	Turn websites, files, uploads, and API responses into normalized documents.
Chunking	`/v1/chunk`	Split documents into retrievable chunks.
Embedding	`/v1/embed`	Generate vectors for chunks.
Storing	`/v1/store`, `/v1/edit`, `/v1/delete`	Upsert, replace, and remove stored chunks.
Retrieval	`/v1/retrieve`	Search and optionally rerank stored chunks.
Chat	`/v1/chat`, `/v1/chat/stream`	Run one-shot or streaming chat.

Streaming

Streaming endpoints return text/event-stream:

Endpoint	Streams
`POST /v1/ingest/url/stream`	URL crawl progress and final scrape result
`POST /v1/chat/stream`	Retrieval/generation progress, tokens, and final chat result

Server Extras

Stack	Install
Qdrant	`pip install "ragrails[server-qdrant]"`
Pinecone	`pip install "ragrails[server-pinecone]"`
Weaviate	`pip install "ragrails[server-weaviate]"`

Most request and response bodies are JSON. Document uploads use multipart form data. Use the OpenAPI schema when generating clients.