RAG Application
Learn how to deploy a retrieval-augmented generation app with Ink.
Create an Ink account and connect Ink
Make sure you have an Ink account and have set up Ink for your agent. Use the CLI, the Ink Skill, or MCP. For MCP auth, choose:
- OAuth — authenticate through a browser flow, no keys to manage
- Token — create an API key at deployink.com/account/api-keys
See Quick Start for full setup instructions.
Verify the connection
Once Ink is connected, confirm your agent can reach it by asking it to check.
Do you have Ink connected?
Build the RAG app
Ask your agent to build a RAG application. The agent will scaffold a React frontend with a Node.js backend that uses LangChain for document processing and Claude for generation.
Build a RAG app that lets users upload documents and ask questions about them. Use LangChain with an in-memory vector store and Claude as the LLM. Build the frontend with React.
Deploy with Ink
Tell your agent to deploy. The app needs an ANTHROPIC_API_KEY environment variable and more memory for embedding operations.
Deploy with Ink. Use my Anthropic API key.
service_create(
"name": "rag-app",
"repo": "ink/rag-app",
"build_pack": "railpack",
"ports": {
"0": "[object Object]"
},
"memory": "512Mi",
"env_vars": {
"0": "[object Object]"
}
)Visit your app
Ink returns a live URL once the build completes. Upload documents and start asking questions.
Check logs and metrics (optional)
Ask your agent to pull build logs, runtime logs, or metrics for your deployed service.
Check metrics and logs
That's it — your RAG app is deployed and live. From here you can add a custom domain, deploy a PostgreSQL database for persistent storage, or explore more examples.