RAGler

RAGler is an open-source RAG knowledge operations platform.

TL;DR

• Ingest knowledge from Confluence, web pages, or manual text.
• Review and edit chunks in a draft session.
• Publish validated chunks to Qdrant collections.
• Query published knowledge from the backend API or MCP server.

Quick Start

Prerequisites

• Node.js 20+
• pnpm
• Docker
• OpenAI API key

1. Start infrastructure

docker compose up -d redis qdrant

2. Start backend

cd backend
pnpm install
cp .env.example .env
# set OPENAI_API_KEY in .env
pnpm start:dev

3. Verify backend

curl http://localhost:3000/api/health/liveness
curl http://localhost:3000/api/health/readiness

Expected liveness response:

{"status":"ok"}

4. (Optional) Start frontend

cd ../frontend
pnpm install
pnpm dev

5. (Optional) Start MCP server

cd ../mcp-server
pnpm install
pnpm build
pnpm start

Configuration

Backend runtime is driven by backend/.env. Core variables:

• OPENAI_API_KEY (required)
• REDIS_HOST (required)
• REDIS_PORT (default 6379)
• QDRANT_URL (required)
• PORT (default 3000)
• SESSION_TTL (default 86400)

Run and Verify Core Flow

1. Create a collection

curl -X POST http://localhost:3000/api/collections \
  -H 'Content-Type: application/json' \
  -H 'X-User-ID: demo@ragler.ai' \
  -d '{"name":"Support KB","description":"Support knowledge"}'

2. Start manual ingest

curl -X POST http://localhost:3000/api/ingest/manual \
  -H 'Content-Type: application/json' \
  -H 'X-User-ID: demo@ragler.ai' \
  -d '{"content":"RAGler stores draft sessions in Redis and publishes to Qdrant."}'

3. Review session and publish

curl http://localhost:3000/api/session/<SESSION_ID> -H 'X-User-ID: demo@ragler.ai'
curl -X POST http://localhost:3000/api/session/<SESSION_ID>/preview -H 'X-User-ID: demo@ragler.ai'
curl -X POST http://localhost:3000/api/session/<SESSION_ID>/publish \
  -H 'Content-Type: application/json' \
  -H 'X-User-ID: demo@ragler.ai' \
  -d '{"targetCollectionId":"<COLLECTION_ID>"}'

Docs

• Product docs: docs/docs
• Specs: specs/
• Backend API Swagger: http://localhost:3000/api/docs

Troubleshooting

• readiness fails: check Redis/Qdrant containers with docker compose ps.
• 401/403 style behavior in clients: confirm X-User-ID header is present.
• Ingest failures: confirm OPENAI_API_KEY and outbound connectivity.

Next Steps

1. Read /Users/alex/ragler/docs/docs/getting-started/installation.md.
2. Walk through /Users/alex/ragler/docs/docs/getting-started/first-collection.md.
3. Review architecture in /Users/alex/ragler/docs/docs/architecture/overview.md.