Skip to main content
You can start building with Silvana in more than one place: launch an existing agent, work with Silvana Book Agent, build custom logic with the SDK, or connect directly through APIs. You don’t need to learn the whole stack first. Start with the path that matches what you want to build and go deeper when you need it.

Onbooard Your Agent

New to Silvana? Start with onboarding to install cloud-agent, create your agent identity, and prepare the initial configuration.

​
Start Building with Silvana

Choose the path that fits your workflow. Each page below takes you directly to the tools and documentation you need.

Launch Your First Agent

Move from setup to a running agent using an existing Agent Space implementation or your own logic.
  • Choose your development path
  • Configure, test, and run the agent

Silvana Book Agent

Learn how the cloud agent works and how it connects trading, RFQ, and settlement workflows.
  • Understand the agent runtime
  • Explore CLI and library capabilities

SDK Intro

Build custom agents with the same Rust libraries Silvana Book Agent uses.
  • Reuse execution components
  • Build custom agent workflows

Agentic API

Connect your application or agent through the Agentic API and start building execution workflows.
  • Set up your integration
  • Move from API access to execution

Agent Sample

Start from a working implementation and see how the main SDK components fit together.
  • Explore a practical example
  • Adapt the flow to your own logic

gRPC API

Explore the services and types behind orderbook, settlement, pricing, and ledger operations.
  • Understand the core gRPC types
  • Navigate the available services

​
Essential Reference

Keep these pages close once you move beyond the first setup.

Configuration

Configure markets, runtime behavior, and agent settings through .env and agent.toml.

Full CLI Reference

Find the available cloud-agent commands, options, and execution flags in one place.

Authentication & Signing

Understand authentication, local signing, and how agent requests are authorized.

Security Model

Learn how transaction verification, local keys, and two-phase signing protect agent execution.