Powered by the Smith runtime

Describe the work. Smith forges it.

Tell Smith how the work gets done. It designs the agents, tools, and routing to do it, then deploys the whole system, live and traced.

No SDK to learn  ·  no infrastructure to manage  ·  deploy in one click

SMITH live
START ORCHESTRATORSupervisoropus-4-8 TOOLKnowledge Base SPECIALISTBillingsonnet-4-6 SPECIALISTTechnicalsonnet-4-6 SPECIALISTRefundshaiku-4-5 RESOLVED

Don't start with agents. Start with the work. One conversation later it's designed, deployed, and traced.

THE FORGE LOOP

Five steps from a sentence to a live system

01

Commission

Describe the work you want done, in plain language.

02

Forge

Smith generates the agent graph, with routing welded in.

03

Inspect

Tune any agent's model, tools, and prompt.

04

Deploy

Compile and publish a versioned live endpoint.

05

Observe

Watch real traffic flow through the graph, traced.

DESCRIBE

Say how the work gets done

No graph code, no SDK. Describe the process you run today and Smith drafts the agents, tools, and routing to run it, showing you the plan as it forges, step by step.

commission
Build a support copilot that triages tickets and issues refunds
Forging your system.
  • Supervisor + 3 specialists
  • Conditional routing
  • Knowledge base tool
  • Compile graph
INSPECT

Every node is yours to tune

Open any agent to change its model, temperature, tools, and system prompt. Smith suggests; you decide. Nothing hides behind one giant prompt.

Supervisor · orchestrator
claude-opus-4.8
deterministic0.2
handoffknowledge-base
Route each customer message to the right specialist. Hold conversation state and decide when the issue is resolved.
DEPLOY

Ship to a live, traced endpoint

One action compiles the graph and publishes a versioned endpoint. Watch real traffic flow through it, with runs, latency, and traces built in.

support-copilot live
smith.run/support-copilot
1,284
runs
1.4s
p50
99.4%
success
ORCHESTRATION

Real topologies, not a smarter chatbot

Most tools stop at a single agent. Smith generates the orchestration: supervisors, specialists, parallel work, and feedback loops.

Supervisor + specialists

A router delegates each task to the right specialist and decides when the job is done.

Sequential pipeline

Each step hands its output to the next, in a fixed order, start to finish.

Parallel fan-out

Run independent agents at once, then aggregate their results into one answer.

Reflective loop

An agent acts, a critic checks, and the work loops until it meets the bar.

WHY SMITH

The graph is the output, not the job

Visual builders hand you a blank canvas; frameworks like LangGraph and CrewAI have you assemble the agents yourself. Smith starts from the work: it generates the graph from your description, then hands you the controls.

01

Legible

The whole system is a graph you can see, never a hidden chain.

02

Editable

Every agent, tool, and edge is inspectable and tunable, before and after deploy.

03

Runnable

One action takes a draft to a live, versioned endpoint on the Smith runtime.

04

Observable

Live traces and metrics are built in, right where you built the system.

FAQ

Questions, answered

How is this different from CrewAI, LangGraph, or a visual builder?

Those tools are agent orchestration first: a framework or canvas where you assemble agents, crews, and graphs yourself. Smith is workflow authoring first: you describe the work, and it generates the multi-agent graph, then lets you tune any node and deploy with tracing, in one place. The graph is an output, not a chore.

Who is Smith for?

People who own a workflow: support leads, ops managers, founders, and developers who would rather skip the scaffolding. If you can describe how tickets get triaged or refunds get approved, you can commission a system that does it.

Do I need to know LangGraph or an agent framework?

No. You describe the work in plain language and Smith generates the graph. The framework details are handled for you, though developers can read and edit the underlying spec.

What actually runs the agents?

The Smith runtime compiles each generated graph into an executable agent system, hosts a versioned endpoint, and streams traces. The studio is the design surface on top of it.

Can I edit what Smith generates?

Yes. Select any agent to change its model, temperature, tools, and system prompt. Smith suggests a model per role; you can override it on any node.

Is it production-ready today?

Smith Studio is in early access. You can explore the full build-to-deploy loop in the live prototype now; real generation and runtime are rolling out in phases.

Stop wiring agents by hand

Describe the work, and watch Smith forge it into a deployed, traced multi-agent system.