Build a knowledge base agent in YAML

There's no built-in telephony or voice channel in Digitorn today, so a real 'voice assistant' isn't something you can ship declaratively yet. What is real and shippable right now: an agent that answers from your own documents instead of guessing, built entirely on the rag module's real tools (ingest, query). If a spoken interface matters to you, this is the piece to build first - the knowledge layer a voice frontend would eventually call into.

The minimum YAML to ship this

1modules:2  rag: {}34agents:5  - id: helper6    modules: [{rag: [query]}]7    brain: { provider: anthropic, model: claude-haiku-4-5, credential: { ref: anthropic_main, scope: per_user, provider: anthropic } }8    system_prompt: |9      Answer using the rag.query tool. Always cite the source.10      If nothing relevant comes back, say so plainly instead of guessing.

Install Digitorn and deploy this agent

# 1. install runtime
curl -sSL https://digitorn.ai/install | sh

# 2. save the YAML above to ~/.digitorn/apps/my-voice-assistant/app.yaml
mkdir -p ~/.digitorn/apps/my-voice-assistant
# paste the YAML into app.yaml

# 3. deploy
digitorn deploy my-voice-assistant